EPA/AA/CTAB/TA/83-6
Technical Report
An Analysis of Carbon Monoxide Emissions As a Function
of Ambient Temperature and Vehicle Parameters
by
Larry C. Landman
July, 1983
NOTICE
Technical Reports do not necessarily represent final EPA de-
cisions or positions. They are intended to present technical
analysis of issues using data which are currently available.
The purpose of the release of such reports is to facilitate the
exchange of technical information and to inform the public of
technical developments which may form the basis for a final EPA
decision, position or regulatory action.
U. S. Environmental Protection Agency
Office of Air, Noise and Radiation
Office of Mobile Sources
Emission Control Technology Division
Control Technology Assessment and Characterization Branch
2565 Plymouth Road
Ann Arbor, Micnigan 48105
-------
-2-
Introduction
Numerous investigators have reported the observations that:
1) Carbon monoxide (CO) exhaust emissions tend to in-
crease as ambient temperature decreases.
2) The "cold-start emissions" (those emitted prior to
light-off of the catalyst) are the most sensitive to a
drop in the ambient temperature [1,5,6,7]*.
This tendency toward increased CO emissions is suspected as
being a major factor causing National Ambient Air Quality
Standard (NAAQS) violations for CO during cold weather condi-
tions [1] .
To investigate possible functional relationships between cold
weather CO emissions and emission control technology or vehicle
parameters, we reviewed existing data on eighteen (18) vehicles
for which cold-start tests were performed in EPA's Controlled
Environmental Test Facility (CETF). These data were generated
between February 1980 and October 1981. Since these vehicles
were not specifically chosen as a representative sampling of
the population of late-model year vehicles, care must be taken
in generalizing the results of these analyses.
Results of these analyses may be used in the development of
future test programs designed to investigate ways to reduce
cold weather CO emissions.
*Bracketed, [ ], numbers indicate references at the end of this
report.
-------
-3-
Conclusions and Reconunendations
In tne limited data base which was used, we conclude that the
most significant vehicle parameter related to CO emissions at
low ambient temperature is the type of fuel metering system.
The vehicles equipped with electronic fuel injection (EFI)
exhibited greater cold weather CO control than did the carbu-
reted vehicles. Of secondary importance is the vehicle's test
weight (ETW) and the type of secondary air injection used.
We had limited success in predicting both cold-start CO emis-
sions and the increase in cold-start CO emissions (in grams per
mile) relative to FTP ambient temperatures (75°F) as functions
of ambient temperature, type of fuel metering system, and ETW.
However, the percent of CO increase (relative to 75°F) corre-
lated poorly with tne available variables. From this, we infer
that an additive rather that multiplicative "correction factor"
for cold weather CO emissions may be a more appropriate form.
We did find a high positive correlation, at low ambient temper-
atures, between the increase in fuel consumed (Bag 1 fuel con-
i
sumption minus Bag 3 fuel consumption) and the corresponding
increase in CO emissions. This is further indication that the
cold-start fuel enrichment systems are responsible for much of
the low ambient temperature CO emissions.
It is difficult to determine, from the data set used for this
report, how important eacn design parameter is, since the data
set does not include every possible combination of all the
parameters. It is obvious, however, that the important para-
meters to concentrate on in future investigations are those
related to the fuel metering, air injection, and cold-start
fuel enrichment systems. In an ideal sense, further cold room
testing would involve test vehicles that had the following
characteristics:
-------
-4-
(1) Vehicles representing all combinations of fuel
metering systems (e.g., I, 2, and 4 barrel carbu-
retion; throttle-bodied fuel injection (TBI); multi-
point fuel injection) and secondary air systems (e.g.,
no air injection, pulse air injection, air pump) used
in late-model vehicles and/or likely to be used in
future vehicles, and '
(2) Vehicles representing various types of cold-start fuel
enrichment systems. (We could then search for readily
identifiable vehicle parameters which correlate with
the cold start effects of these systems, and quantify
the effects on cold weather CO emissions.)
Test Vehicles & Test Results
The 18 test vehicles are described in Table 1. As indicated in
that table, data from the first ten (10) vehicles were used in
reference 1, the next five (5) appeared in reference 2, and the
last vehicle appeared in reference 3. Data generated by vehi-
n >
cle D160 (1981 Chrysler K-Car Wagon) and vehicle 1009 (Ethyl's
lean-burn concept car) have not appeared in any previous EPA
technical report.
It should be noted that both tne Ethyl lean-burn car (a 1978
Ford Fairmont with a modified Peugeot V-6 engine and an Asian-
Warner A4(00) transmission) and the Dresser Nova (a modified
1977 Chevrolet Nova equipped with a Dresser sonic carburetor)
are prototype vehicles which embody hardware that is not used
on production cars. Thus, some of the regression analyses were
done both with and without these two vehicles. Additionally,
vehicle FB289 (the turbocharged Datsun 280ZX) was an experi-
mental prototype and, thus, not necessarily representative of
either the 1981 or 1982 model year Nissan vehicles.
-------
-5-
Table 1
Test Vehicle Descriptions
1
2
3
4
5
6
7
8
9
10
1 1
12
13
14
15
16
17
18
MFR
Ford
Nissan
Nissan
GM
GM
Ford
GM
Nissan
Plymouth
Toyota
Chry
Ford
Ford
GM
GM
Chry
Ethyl
Dresser
VI D
9S1-5.8M-H-400
FB223
FB178
P0948
03H2-94472C
8R10Y131366
4M47AAH202725
FB289
D162
MA46100183
B3BK26B4BC184143
1FABP082XBW20356
1FABP21B3BK15310
1G1AX68X7B621755
2G1AW69J63144479
D160
1009
1X69L7L104103
YEAR/
STANDARD
1979 CA
1980 49
1980 CA
1981 49
1980 CA
1978 CA
1980 50
N/A
1981 49
1980 50
1981
1981
1981
1981
1981
1981
N/A
N/A
MODEL
T-BIRD
280ZX
280ZX
GRAND PRIX
CUTLASS SUPREME
PINTO
REGAL
280ZX TURBO
RELIANT
CELICA SUPRA
K-CAR
ESCORT WAGON
FAIRMONT
CITATION
MALIBU
K-CAR (WGN)
ETHYL LEAN -BURN
DRESSER ' 77 NOVA
ciotr, /
No. 'of Cyl
351(5. 8)/8
168(2. 8)/6
168(2.8)/6
265(4. 3)/8
260(4.3)/8
140(2. 3)/4
231(3. 8)/6
168(2. 8)/6
135(2. 2)/4
156(2.6)76
135(2. 2)/4
98(1. 6)/4
200(3. 3)/6
173(2. 8)/6
267(4.4)/S
135(2. 2)/4
163(2.6)/6
350(5.7)/8
-------
-6-
Table 1 (Continued)
1
2
3
4
5
6
7
8
9
10
1 1
12
13
14
1 D
16
17
18
ETW
(lb)
4500
3250
3125
3875
3500
2750
3750
3250
2750
3000
2750
2500
3375
3000
3625
2625
2750
4000
TRAN
A3
A3
M5
L3
A3
A3
A3
A3
M4
A4-OD
A3
M4
A3
A3
A3
M4
A4-OD
A3
SGR
YES
YES
NO
YES
YES
YES
NO
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
Test Vei
- EMISSK
AIR
PUMP
PULSE
NO AIR
PUMP
PUMP
PUMP
PUMP
NO AIR
PUMP
NO AIR
PUMP
PUMP
PUMP
PULSE
PUMP
PUMP
PUMP
NO AIR
hicle Descriptic
DN CONTROL SYST!
CATALYST *
OXD(M)
OXD
3WY
3WY
3WY
3WY
3WY
3WY
3WY
3WY
3WY
(M)
(M)
( P ) +OXD ( P )
(P)
(M)+OXD(M)
(P)
(M)
(M)+OXD(M)
(M)+3WY(P)
(M)+OXD(M)
3WY(M)+OXD(M)
3WY
3WY
3WY
3WY
(M.)+OXD(M)
(M)+OXD(M)
( P ) +OXD ( P )
(M)+OXD(M)
NONE
3WY
(P)
3ns
LM
A/F
OPEN
OPEN
CLOSED
CLOSED
CLOSED
CLOSED
CLOSED
CLOSED
CLOSED
CLOSED
CLOSED
OPEN
OPEN
CLOSED
CLOSED
CLOSED
OPEN
OPEN
FUEL
METR.
2
bbl
EFI
EF
2
2
2
2
EF
2
T
bbl
bbl
bbl
bbl
I
bbl
EFI
2
2
1
2
2
2
3
bbl
bbl
bbl
bbl
bbl
bbl
bbl
SONIC
REF
[1
[1
[1
[1
[1
[1
[1
[1
[1
[1
[2
[2
[2
[2
[2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
]
None
[4
[3
3
3
? 'P' designates a Pelleted catalytic converter,
and 'M' designates a Monolith.
-------
-7-
The CO exhaust emissions from those test vehicles/ over the FTP
driving cycle, appear in Appendix 1. The emissions, in grams
per mile, are given for the cold start (Bag 1) , hot transient
(Bag 2) , and the hot start (Bag 3) portions of the FTP cycle.
Those data are presented graphically, for each vehicle, in
Appendix 2.
Data Analyses
As the 18 graphs in Appendix 2 illustrate, the CO emissions
during the hot transient (Bag 2) and hot start (Bag 3) portions
of the FTP are relatively low and relatively insensitive to the
ambient temperature. The only possible exception to this
observation could be the performance of three of the four Ford
vehicles (see Figures 07, 12, and 13 of Appendix 2) . These
three vehicles exhibit a trend in low temperature Bag 2 and Bag
3 CO emissions that appear somewhat different from the other
vehicles. For the purposes of this report, no special analysis
based on this observation was made. However, it may be neces-
sary and appropriate to consider the low temperature Bag 2 and
Bag 3 results of vehicles in future test programs, based on
this observation.
For this report, the. analysis was divided into the following
five areas:
1) Analysis of the cold start (Bag 1) CO emissions only.
This results in analysis of Bag 1 emissions in grams
per mile.
2) Analysis of the amount by which the Bag 1 CO emissions
exceeded the Bag 3 CO emissions (i.e., the difference
of Bag .1 and Bag 3 emissions). This results in analy-
sis of grams per mile differences.
-------
-8-
3) Analysis of the amount by which the Bag 1 CO emissions
exceeded the average Bag 1 CO emissions at 75°F for
each vehicle. This results in analysis of grams per
mile differences.
4) Analysis of the ratio of the Bag 1 CO emissions to the
average Bag 1 CO emissions at 75°F. This results in
analysis of dimensionless quantities equivalent to
percent increases in CO emissions.
5) Analysis of the ratio of the Bag 1 CO emissions to the
corresponding Bag 3 emissions. This results in analy-
sis of dimensionless quantities equivalent to percent
increase in CO emissions.
For each vehicle, equations were generated to approximate CO
emissions by using a "best fit" first-degree and a "best fit"
second-degree polynominal in ambient temperature (Appendix 3) .
The best fit was obtained using the least-squares-fit method,
which smoothed the data, and provided an insight into the shape
of the CO emissions versus ambient temperature curve. A second
set of equations was generated by using results from tests for
which the ambient temperature was below 90°F. Selection of a
quadratic equation in ambient temperature to model CO emissions
was based on a visual observation of the data. That is, the
quadratic is tne simplest function that is concave upwards and
decreases as temperature increases. In addition, modeling
the CO emissions with such a function has been used in the
past. [5] For each of the 64 least-squares-fit equations for
Bag 1 CO in Appendix 3, we calculated:
1) predicted CO emissions at 20°, 40°, and 75°F,
-------
-9-
2) predicted slope of the equation at 20°, 40°, and 75°F,
3) predicted sensitivity of the equation at 20°, 40°, and
75°F, where sensitivity is the per cent change in CO
emissions divided by the per cent change in temper-
ature,
4) ratio of the predicted CO emissions at 20°F to those
at 75°F.
For each of the four combinations of temperature interval and
polynominal degree/ we computed correlation coefficients
between each of those calculated values and the following
eleven (11) vehicle parameters:
1) secondary air (air pump, pulse air, or no air)
2) number of catalysts (0, 1, or 2)
3) equipped with an oxidation catalyst only (yes or no)
•. i
4) equipped with a three-way catalyst only (yes or no)
5) .. equipped with both a three-way and oxidation catalyst
(yes or no)
6) type of catalyst (pelletted or monolith)
7) fuel metering system (fuel injected or carbureted)
8) control of air/fuel ratio (open or closed loop)
9) equipped with EGR (yes or no)
-------
-10-
10) equivalent test weight (ETW) (in pounds)
11) number of cylinders (4, 6, or 8)
Conspicuous by its absence from the preceding list is a vehicle
design parameter directly relating to cold-start fuel enrich-
ment (e.g., choke angle as functions of time and temperature).
Thus, this important factor was omitted from the list because
no variable was found to consistently represent it for each of
the test vehicles. However, some of the characteristics of
cold-start fuel enrichment may be inferred from the amount of
additional fuel consumed to perform the cold-start relative to
the corresponding hot-start (i.e., Bag 1 minus Bag 3 fuel con-
sumption) . This difference in total fuel consumed (in grams)
appears in Appendix 1.
The matrices of the calculated correlation coefficients appear
in Appendix 4. Each correlation coefficient, r, between pairs
of variables is the standard product-moment correlation coef-
ficient.* This value is used to describe the strength of the
linear relationship (if any) between the two variables being
correlated. The tables in Appendix 4 give the correlation
coefficients between pairs of actual data (from Appendix 1) and
vehicle parameters, between pairs of model predicted data (from
Appendix 3) and vehicle parameters, and between pairs of vehi-
cle parameters. The selection of model predicted values was
made to allow us to correlate the shape of the CO values was
made to allow us to correlate the shape of the CO emission
curves (i.e., slope and sensitivity) with the various vehicle
parameters.
*For example, see Introduction to the Theory of Statistics by
Yale, G.U., and Kendall; M.G. Hafner Publishing Company, New
York 1968, p. 216.
-------
-11-
Choice of numerical values on which to base a judgment about
whether or not there is a relationship of the type being in-
vestigated is always a matter of choice. For this analysis we
judged the variables to be linearly related, if we can accept
that null hypothesis of linear dependence at the 0.01 signifi-
cance level. The only vehicle parameters thus defined which
correlated highly (either positively or negatively) with the
low temperature CO values (either actual or calculated) were
those related to the fuel metering system, secondary air
system, ETW, and possession of only a three-way catalyst. Un-
expectedly, the ratio of the CO emissions at 2U°F to CO emis-
sions at 75°F (or equivalently, the percent of increase in the
CO emissions at lower temperatures) did not correlate highly
with any of those eleven vehicle parameters. Possession of
only a three-way catalyst was dropped as a significant variable
because of its high correlation with secondary air systems.
The increase in Bag 1 relative to Bag 3 CO emissions (i.e., Bag
1 minus Bag 3 emissions) correlated highly with the corre-
sponding increase in fuel consumption at ambient temperatures
below the FTP temperature range (68°F to 86°F). (Appendix 4)
It, therefore, appears likely that the temperature-related in-
creases in CO emissions are due to the cold-start fuel enrich-
ment system.
We tnen used both stepwise forward and stepwise backward re-
gression analyses to model the actual cold-start CO emissions
as a function of the three (3) continous variables: TEMP
(ambient temperature), TEMP.2 (square of the ambient temper-
ature) , and ETW. The TEMP variables were chosen because of the
success in modeling individual car CO results (see Appendix
3) . ETW was chosen because, in the correlation matrices in
Appendix 4, ETW was almost always the next most highly corre-
lated variable in the linear sense after the three following
-------
-12-
categorical variables which were previously determined to be
(linearly) related to cold weather CO at the 0.01 level of
significance:
PMP (0 = No air pump, 1 = Has air pump)
PLS (0 = No pulse air, 1 = Has pulse air)
FS (0 = EFI, 1 = Carbureted)
i
It is important to note that, in this sample of 16 cars, those
six variables are not independent. The variables PMP and PLS
are, of course, always mutually exclusive (i.e., no car is
equipped with both an air pump system and a pulse air injection
system).. However, in this study, the type of secondary air in-
jection system (PMP or PLS) correlates highly with the type of
fuel metering system (FS) : all the test vehicles with air
pumps were carbureted, while all the vehicles without secondary
air were fuel injected. Thus, for this sample of 16 cars, the
combinations of the categorical variables produce only four,
rather than six, subsets with data in them.
,!l
S.tepwise forward regression analyses were performed using those
variables, products of those variables with the categorical
variables, and with products of the categorical variables.
Stepwise backward regression analyses had to be performed with
a reduced set of variables because the matrix of coefficients
used by the. MIDAS statistical program became singular when all
the variable and'product terms were used.
The end results of those regression analyses are given in
Appendix 5. Those results indicate that the CO emissions can
be modeled using those variables, and those models yield
multiple-Rs of aoout 88 percent for the vehicles in this
sample. In light of the high degree of correlation among what
should be independent variables, the final regression equations
-------
-13-
are not vital. What is important, however, is the fact that
the most critical variable ('after, "temperature") in modeling
cold-start CO emissions is the fuel metering system (FS) and
possibly air systems. This observation is illustrated in
Figures 1 to 3.
Equally important is the fact that attempts to model the ratio
of increase in CO emissions from 75°F resulted in multiple Rs
of 0.66 to 0.68. Thus, the proportion of the variability (the
square of R) in that ratio of CO emissions that is explained by
temperature, ETW, and the categorical variables is less than 46
percent. This seems to indicate that use of a multiplicative
form of "correction factor" to calculate cold weather CO may
not be as good as an additive form.
If we use tne results of the regression analyses in Appendix 5,
not to predict CO emissions, but rather to predict the differ-
ence in CO emissions between the carbureted and fuel injected
vehicles at low ambient temperatures (i.e., 20°F), the
equations predict:
1) Cold-start CO emissions of the carbureted vehicles
exceed those of the fuel injected vehicles by 72 grams
per mile. (The actual data indicate 79.6 grams per
mile.)
2) The increase in cold-start CO emissions from 75°F to
20°F for the carbureted vehicles exceeded by 65 grams
per mile the corresponding increase for the fuel in-
jected vehicles. (The actual data indicate 74.2 grams
per mile.)
These predictions are borne out by Figures 1 and 2.
-------
-14-
F i qure 1
Average Bag 1 CO Emissions Versus Ambient Temperature
(Means of Actual Data from Appendix 1)
Stratified by Fuel Metering & Secondary Air Systems
CO Emissions
(g/mi)
1 10
100 *
X
so
50
40
30
20
1C
EF1 + Pulse
20 25 30 35 40
45 50 55 SO S5 70
Ambient Temperature (*F)
75 8O 85 9O 95 100
-------
CO Emissions
(g/mi)
90+4
-15-
Fi qure 2
Average Difference between Bag 1 CO Emissions and
Bag 1 CO Emissions at 75'F Ambient Temperature
(Means of Actual Data from Appendix 1)
Versus Ambient Temperature
Stratified oy Fuel Metering & Secondary Air Systems
80
70
Garb + Air Pump
SO
5O
10
EFI + Pulse Air
ALL
-10
20 25 30 35 40 45 50 55 60 55 70 75 80 85 90 95
Ambient Temperature ('F)
-------
-16-
Flqure 3
Average Difference between Bag 1 CO Emissions and
Bag 3 CO Emissions Versus Ambient Temperature
(Means of Actual Data from Appendix 1 )•
Stratified by Fuel Metering & Secondary Air Systems
CO Em i ss i ons
(g/mi)
•;00
90
80 +
70
SO
-i- 3
50
40 •*•
30
20
10
Carb + Air Pump
EFT -i- Pulse Air
20 25 30 35 40 45 50 55 SO S5 70 7" 80 85 90 95 100
Ambient Temperature (*F)
-------
-17-
References
1. Robert I. Bruetsch, "Carbon Monoxide and Non-FTP Ambient
Temperature," EPA Technical Report Number EPA/AA/CTAB/ TA/
81-7, February 1981.
2. Rodney J. Branham, "Evaluation of the Temperature Effects
on Five 1981 Passenger Vehicles," EPA Technical Report
Number EPA-AA-TEB-82-4, December 1981.
3. Thomas J. Penninga, "Evaluation of a Dresserator System
Test Vehicle," EPA Technical Report Number EPA-AA-TEB81-21,
June 1981.
4. F.J. Marsee, "Ethyl Concept Car vs. 1980 Production Car,"
Ethyl Corporation Research Laboratories, Report Number ERM
80-4, May 1980.
5. N. Ostrouchov, "Effect of Cold Weather on Motor Vehicle
Emissions and Fuel Economy," SAE Paper Number 780084,
HI
February-March 1978. . ,;
6. R.S. Spindt and F.P. Hutchins, "The Effect of Ambient
Temperature Variation on Emissions and Fuel Economy - An
Interim Report," SAE Paper Number 790228, February-March
1979. .
7. J.N. Braddock, "Impact of Low Ambient Temperature on 3-Way
Catalyst Car Emissions," SAE Paper Number 810280, February
1981.
-------
APPENDIX 1
CO Emission Data from FTP Driving Cycle
1-1
-------
Appendix 1
CO Emissions Data from FTP Driving Cycle
Vehicle I .D.
9S1-5.8M-H-400
1979 fold T-8-Ud
FB223
J9S0 49-S;tate
VcutAun 2SOZX
FB178
1980 Ca£x,|$0AnA.a.
PajUun 2S0ZX
P0948
1981 GJiand P/U.X.
03H2--94472C
J9S0 Cu;t£a4.6 Sap
Temp
(°F)
20.0
60.0
60.0
74.0
75.0
75.0
100.0
101.0
21 .0
22.0
60.0
60.0
74.0
75.0
75.0
75.0
99.0
100.0
100.0
20.0
20.0
60.0
60.0
75.0
75.0
75.0
75.0
20.0
20.0
60.0
60.0
75.0
75.0
1 00.0
101.0
20.0
20.0
60.0
62.0
74.8
75.0
75.0
100.0
100.0
Test
Number
801423
801424
801427
801454
801632
801648
801579
801580
802513
802514
801642
801644
801641
801640
802414
802578
801639
801643
802413
802687
802810
802586
802589
802434
802435
802681
802684
803474
803476
803098
803101
803092
803095
803404
80.3479
803783
803786
803597
803590
803777
803593
803776
803779
803781
CO Emissions (g/mi)
Bag 1
201.919
105.835 •
106.355
46.897
47.816
53.442
13.369
10.528
22.348
19.172
7.280
7. 173
6.219
9.375
6.253
8.661
. 9.571
8.088
11.101
30.701
27.514
10.797
10.375
1 1 . 102
7.010
13.953
7.666
141 . 126
159.249
77.532
69.041
9.597
10.308
5. 162
3.604
98.346
107.343
39.290
38.939
10.487
12.910
14.462
7.305
6.653
Baq 2
4.318
0.601
2.201
0.632
0.921
0.709
3.259
3. 127
0.0
0.0
-0.007
0.019
0.019
0.014
0.012
0.005
0.025
0.029
0.0
1 .239
1.342
0.960
1.042
1 .044
1 . 143
1 .250
0.737
0.221
0.096
0.1 16
0. 106
0.321
0. 184
0.096
0.062
0.702
0.400
0.326
0.323
0.229
0.284
0.314
0.649
0.275
Bag 3
9.989
9.409
7.220
7.761
9.352
5.725
20.331
13.388
o!.072
0.537
0.567
0.019
1.569
0.513
1.283
0.969
1 .444
1.270
2.360
1 .520
1 .434
0.966
0.901
0.871
0.852
1 .342
1,201
:; i
0'>.812
1. 189
0.678
1.030
2.803
3.060
1 .784
0.709
0.648
0.655
0.622
0.892
0.590
0.658
0.849
2.41 1
1 .266
AFuel
Used*
313. 1
227.5
194.6
158.4
151 .3
170.4
84.7
59.0
202.6
173.4
1 14.5
126.6
34.9
90.8
203.5
41 .2
55.6
60.7
50.4
173.5
164.0
97.1
97.6
87.4
77.6
97.8
75.2
441 .6
456.8
272.4
246.5
132.0
130.0
92.0
89.3
265.2
271 .3
139.5
126.9
96.6
96.5
90.9
27.0
36.3
* AFuel Used is the amount of fuel (in grams) consumed during Bag
minus the amount consumed during Bag 3. (Fuel consumption was
using
calculated
carbon balance equations.)
-------
1-3
'Appendix 1 (Continued)
CO Emissions Data from FTP Driving Cycle
Vehicle I .D.
8R10Y131366
J97S Pon.d Pj.n£o
4M47AAH202725
19 SO BiUcfe Re.ga£
FB289
19Z1 .5 fl
-------
1-4
Appendix 1 (Continued)
CO Emissions Data from FTP Driving Cycle
Vehicle I .D.
MA46100183
19 SO Ce.6cc.fl. Sup/to.
B3BK26B4BC184143
J9SJ Ckx.y K-CO.A
1FABP082XBW20356
19S1 EtcoJit Wagon
1FABP21B3BK15310
J9SJ Fcu,A.mon£
1G1AX68X7B621755
J9SJ CLtaJtcon
2G1AW69J6B144479
1931 Cnev Ma-Lcfau
D160
I93J CJuiy K-Wagon
Temp
(°F)
20.0
20.0
60.0
60.0
71.5
72.5
73.0
74.5
74.5
75.0
75.0
75.0
75.0
100.0
100.0
22.0
59.0
72.5
97.0
23.0
58.5
75.0
98.0
22.2
59.0
74.0
100.0
22.0
60.5
75.0
101 .5
18.7
60.0
73.6
20.0
60.0
75.0
75.0
100.0
100.0
Test
Number
806710
806713
806704
806707
805387 *
805537
806359
805383
806357
805385
806361
806698
806701
806690
806693
810713
81071 1
810697
810734
810849
810826
810822
810824
810575
810573
810569
810571
810752
810750
810746
810748
81 1 194
81 1188
81 1 180
807526
807528
807389
807392
807381
807386
CO Emi
Baa 1
36.404
36. 169
10.590
13.954
70.073
6.873
6.685
5.234
6.490
6.887
5.440
10.784
5.987
3. 137
3. 163
125.399
23.944
22.938
12.355
78.402
14. 142
11 .206
10.620
136.083
3.578
3.980
3.853
58.147
23.018
.17. 174
1 1 .036
133.036
8.836
9.080
97.453
21 .637
12.710
13.355
7.043
6.646
ssions
Baq 2
0.486
0.469
0.436
0.490
61.705
0.217
'0. 113
;
-------
1-5
Appendix 1 (Continued)
CO Emissions Data from FTP Driving Cycle
Vehicle I.D.
1009
Le.o.n-Bu
-------
APPENDIX 2
Plots of CO Emissions for Each Test Vehicle
(Bag-by-Bag)
(Note: All CO scales are not the same.),
2-1
-------
2-2
Figure 01
Scatter Plot of Bag-1 (1), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. 9S1-5.8M-H-4OO, Ford 1979 California T-BIRD (351 cid)
CO Emissions
(si/mi)
225
200
175
150
125
1
1
100 +
75
1
50 +
1 1
25 +
3
+ 13
3 33 33
2 23 22
0 + 2. 222
+.— _ — _.*_..._ ...«.•..._ — .^—-~_.^.. —_—^__-_ + — _«_.*— __«^.._*^ —— __ ^. —~.^_...^« M..^...«4..«. .«^.v_^._. __.*.____ *
15 20 25 30 35 40 45 50 55 60 65 70 75 8O 85 90 95 1OO 105
Ambient Temperature (*F)
-------
2-3
Figure 02
Scatter Plot o-f Bag-1 (1), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. PB223, Nissan 1980 49-State 280ZX (168 cid)
CO Emissions
(g/mi)
40
35
30
25
20 +
1
15
1
10 +
1 1
1 1
* 1 1
+ 3
33 33
3 3 223
+ 32 32 22 222
4,_......4...__..4._ _4> — __-4..»«__4._ .*.«.-- — ••«>•• — -- + — — — — 4- — — *— 4- •-— •*• — — — — -*• — —+— — —-4-—— -—4» — —— —4- — — — —4>
15 20 25 30 35 40 45 50 55 SO 65 70 75 80 85 90 95 100 105
Ambient Temperature (*P)
-------
2-4
F1qure 03
Scatter Plot of Bag-1 (1), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. FB178, Nissan 1980 California 280ZX (168 cid)
CO Emissions
(g/mi)
45
40
35
1
30 +
25
20
15
1
10 •»• 1
33 2
2 33 333
0 +
15 20 25 30 35 40 45 50 55 SO 65 70 75 SO 85 90 95 1OO 105
Ambient Temperature CF)
-------
2-5
Figure 04
Scatter Plot of Bag-1 (1), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. P0948. GM 1981 49-State GRAND PRIX (265 cid)
CO Emissions
(g/mi)
180
160
140
120
100
80 +
1
+ 1
SO
40
20 +
+ 1 1
1
33 31
0 •*• 23 32 22 ' 22
+ ___ _ + ----.*----•»•- _--<.- ___+____+____*- _4.--__4.----*----*----J-----4.-- 4-- -f- + - - —
15 20 25 30 35 40 45 50 55 SO 65 70 75 80 85 30 95 100 105
Ambient Temperature (*F)
-------
. Figure OS
Scatter* Plot of Bag-1 (Di Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. 03H2-94472C, GM 1980 California CUTLASS SUPREME (260 Cid)
CCi Emissions
(ei/mi)
135
120
1
105 +
90
75
SO
•*• 1
45 *
1 1
3O
15 * 1
1
1
* 11
33
0 * 22 2 3 333 22
4.- _.,._ _-_-!. _+.____* __ + ____+ + ____4. __*____* ---«.- _ + __--*-..__ + __-._ + ____^____^.____4.
15 20 25 30 35 40 45 50 55 SO 65 70 75 3O 35 9O 95 100 105
Ambient Temperature CF)
-------
2-7
Figure 06
Scatter Plot of Bag-1 (1), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. 8R10Y1313B6, Ford 1978 California PINTO (140 cid)
CO Emissions
(g/mi)
135
12O
105
90 +
1
75
SO
45
30 +
1 1
+ 3
3
15 +
22 1
33
0 + 33 23 2
4.____4._'_ 4.____4--___J.-__—. ____4.____^.___ _j.__ __.____*____*__ __^.__ __ + ____+_-_-*----»_---*-_-_•
15 20 25 30 35 40 45 5o' 55 SO 65 70 75 80 85 90 95 100 105
Ambient Temperature (*F)
-------
2-8
Figure 07
Scatter Plot of Bag-1 (1), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. 4M47AAH202725, GM 1980 50-State REGAL (231 Cid)
CO Emissions
(g/mi)
90
80
70
60 +
1
+
50 +
•*•
40 +
•«•
30 +
1
20 +
1
10 + 3
3
23 2
-33 23 2
0 +
15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105
Ambient Temperature (*F)
-------
2-9
Figure 08
Scatter Plot of Bag-1 (1), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. FB289, Nissan 1981.5 49-State 280ZX TURBO 1168 cid)
CO Emissions
(g/mi)
45
40
35
3O
25
20
15
10
5 11. 1
1 1
1
+ 1
233 23
223 223 22 33
0 +
15 20 25 30 35 40 45 50 55 SO 65 70 75 80 85 90 95 1OO 105
Ambient Temperature ('F)
-------
2-10
Figure 09
Scatter Plot of Bag-1 (1), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. D162. Plymouth 1981 49-State RELIANT (135 cid)
CD Emissions
(9/mi)
135
120
105
90
75
60
45 *
30
15 + 11
1
+ 1 111
33 33 33 33
0+22 22 22 32
+ _ — __ + — — _—* — — — • + .. — — -.4.- — «-.+.-...-_ + _.. — •.*._ _ _ — •*• — - — + — — — — + — — — _ + ___ .^_^*_ ^« .«.^....^....^._..^....4.
15 20 25 30 35 40 45 50 55 SO 65 70 75 SO 85 90 95 100 105
Ambient Temperature (*F)
-------
2-11
Figure 10
Scatter Plot o-f Bag-1 (1), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. MA46100183, Toyota 1980 50-State CELICA SUPRA (156 cid)
CO Emissions
(g/mi)
40
1
1
35 +
30
25
20
15
10
111
1 1
1
1 1
22 22 2223 3
0 + 33 2222 22
•.f...___4.__ __.*___...*_ — 4. — ~* — — •* — — -.—4- — - - —* — — - — + — - — — •*• — — — — -f — — --"f- — — *-—— — ^.- — -fc — — - -« ^ — * — — — — J.
15 20 25 30 35 40 45 50 55 SO 65 70 75 80 85 90 95 100 105
Ambient Temperature (*F)
-------
2-12
Figure 11
Scatter Plot of Bag-1 M), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. B3BK2684BC184143, Chrysler 1981 K-CAR (135 Cid)
CO Emissions
(g/mi)
135
120
105
90
75
60
45 ,'
30
15'
3 3.
3 3
0+2 22 2
^— __...t.___..^__._.. + ~..__ + ..-._-..t_ — — « + —_ __.+.— ___.*•_ «_^._.«^_. ..^.....^....^*.._^....^. _ + _ _^_..«4.
15 20 25 30 35 40 45 50 55 SO 65 70 75 3O 85 SO 95 100 105
Ambient Temperature (*F)
-------
2-13
Figure 12
Scatter Plot of Bag-1 (1), Bag-2.(2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. 1FABPO82XBW20356, Ford 1981 ESCORT WAGON (98 Cid)
CO Emissions
(g/mi)
9O
80
70
SO
50
40
30
20 +
3
4>
1
1
10 + ' 1
3
* 3
3
2 22
0 + 2
+ ____*____*__-_* -4.-___-i. __ + ____4._ ___*_-__ + ____*--__ + _-_-«._-_- — ----+----* * -- + - t.
15 20 25 3O 35 40 45 50 55 SO 65 70 75 SO 85 9O 95 10O 105
Ambient Temperature ('F)
-------
2-14
Figure 13
Scatter Plot of Bag-1 (1), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. 1FABP21B38K15310, Ford 1981 FAIRMONT (20O cid)
a) Emissions
(jj/mi)
180
1SO
14O
120
10O
80
SO
20
1 1 3
32 2
15 20 25 3O 35 40 45 50 55 SO 65 70 75 80 85 90 95 100 105
Ambient Temperature CF)
-------
2-15
F i qure 14
Scatter Plot of Bag-1 (V), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. 1G1AX68X7B621755, GM 1981 CITATION (173 cid)
CO Emissions
(g/mi)
90
SO
70
SO
50
40 •*•
30
20 *
1C •*•
3 333
0+2 22 2
+ __ .k_ — .^..A.^..._ _+_ __'_+.. ...^.. —^..«_4.....^._«.~^.~._«^«.«.^ —— — .^. — -•^—.-—^—- ——^- — — _+_- — 4.
15 20 25 3O 35 40 45 50 55 SO 65 70 75 80 85 90 95 10O 1O5
Ambient Temperature ('F)
-------
2-16
Figure 15
Scatter Plot of Bag-1 (1), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. 2G1AW6906B144479, GM 1981 MALIBU (267 Cid)
CO Emissions
(g/mi)
180
160
140
120
10O
80
SO
40
20
3 3 ' 3
2 22
15 20 25 30 35 40 45 50 55 SO 65 70 75 30 85 9O 95 100 105
Ambient Temperature (*F)
-------
2-17
Figure 16
Scatter Plot of Bag-1 (1), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. 0160, Chrysler 1981 K-CAR (WON) (135 eid)
CO Emissions
(g/mi)
135
120
1O5
90
75
60
45
30
15 +
1 1
11
3 3 33 33
+ 2 2 22 22
+ _ _.*.____ + ___-. + _ h- - + — — •*—-. — — .». _^_«_-^« — -—^——-^^— ---^, —» —— ^«_ ——^. — -._^.—— .—^ —. ——^__ —— ^.
15 2O 25 30 35 40 45 50 55 60 65 70 75 3O 85 90 95 100 105
Ambient Temperature ('F)
-------
2-18
Figure 17
Scatter Plot of Bag-1 (1), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. 1009, Ethyl Lean-Burn Car (163 cid)
CO Emissions
(g/rni)
180
160 + 1
1
+• 1
140 +
120
100
80
1
SO * 1
40
20 + 11
1 1 1
1
+ 33 333
333 3 333 33 233
222 2 222 222 2 2222
0 +
15 20 25 30 35 40 45 50 55 SO 65 70 75 80 85 90 95 100 105
Ambient Temperature C'F)
-------
2-19
Figure 18
Scatter Plot of Bag-1 (1), Bag-2 (2), Bag-3 (3) CO Emissions versus Ambient Temperature
Vehicle No. 1XB9L7L104103, Dresser 1977 NOVA (350 cid)
CO Emissions
(g/mi)
45
40
35
30
25
20
15
10
1 1
1
1 31
1 1
1 1 1
222 11
23 22 22
33 222 3 222
22 2
15 20 25 30 35 40 45 50 55 SO S5 70 75 30 35 90 95 100 105
Ambient Temperature ("F)
-------
APPENDIX 3
Equations Modeling CO Emissions for Each Vehicle
3-1
-------
3-2
Models Predicting Cold-Start CO Emissions for Vehicle 9S1-5.8M-H-400,
Ford 1379 California T-BIRO (351 cid)
N Temp interval
8 20.0 - 101 .0
a 20.0 - 101 .0
8 20.0 - 101 .0
a 20.0 - 101 .0
6
6
6
6
Models
N
1 1
1 1
1 1
1 1
8
3
8
a
Models
N
"3
8
a
8
Models
N
a
a
8
8
6
6
e
6
20.0 - 75.0
2O. 0 - 75.0
2O. 0 - 75.0
2O. 0 - 75.0
Predicting Co
Temp Interval
21.0 - 100.0
21.0- 100.0
21.0 - 1 00 . 0
21.0 - 100.0
21.0- 75.0
21.0- 75.0
21.0- 75.0
21 .0 - 75.0
Predicting Co
Temp Interval
2O. 0 - 75.0
20.0 - 75.0
2O. 0 - 75.0
20.0 - 75.0
; Predicting Co
Tamp Interval
2O. 0 - 1O1 .0
2O. 0 - 101 .0
20.0 - 101 .0
20.0 - 101 .0
2O. 0 - 75.0
20.0 - 75.0
20 . 0 - 75 . 0
20.0 - 75.0
BAG1
BAG1
(BAG1
(BAG1
BAG1
BAG1
(BAG1
(BAG1
CO
CO
- BAG3)
- BAG3)
CO
CO
- BAG3 )
- BAG3)
S
s
CO =
CO =
3
3
CO =
CO «
243.77
271 .94
239.69
255 . O4
263.44
218.47
252.55
207 . 80
- 2.4142(TEMP)
- 3.5O45(TEMP)
- 2.5036(TEMP)
- 3. 0978 (TEMP)
- 2.7977(TEMP)
- 0.3004 (TEMP)
- 2. 7541 (TEMP)
- 0.2685(TEMP)
Id-Start CO Emissions for Vehicle FB223,
! .
BAG1
8AG1
(BAG1
(BAG1
BAG1
BAG1
(BAG1
(BAG1
CO
CO
- BAGS)
- BAG3)
CO
CO
- BAGS)
- 3AG3)
3
S
CO =
CO =
3
3
CO =
CO =
20. 14
33.63
2O.49
33.20
25.33
37.97
25.44
36.49
Id- Start CO Emissions for
BAG1
BAG1
(BAG1
(BAG1
CO
CO
- BAG3)
- BAG3)
3
3
CO =
CO =
35.26
47.52
33.68
45.28
'Id-Start CO Emissions for
BAG1
BAG1
(BAG1
(BAG1
BAG1
BAG1
(BAG1
(BAG1
CO
CO
- BAG3)
- BAGS)
CO
CO
- BAG3)
- BAG3)
3
3
CO -
CO =»
8
8
CO =
CO =
183.30
216. 13
182.37
216. 19
202 . 36
138.46
2O2 . 67
134.29
- 0. 1 397 f TEMP)
- 0.7044 (TEMP)
- 0. 1586(TEMP)
- O.S910(TEMP)
- 0.2514(TEMP)
- 0.9578(TEMP)
- 0.2S53(TEMP)
- 0.8832(T£MP)
Vehicle FB178,
- 0.3543(TEMP)
- 1.0732(TEMP)
- 0.34S5(TEMP)
- 1.0270(TEMP)
Vehicle P0948,
- 1.9388 (TEMP)
- 3.4312(T)^MP)
- 1.9479 (TEMP)
- 3.4853(TEMP)
- 2.42O4(TEMP)
+ 1.4226(TEMP)
- 2.4475(TEMP)
+ 1.S330(TEMP)
+ 0.
+ 0.
- 0.
- 0.
008775 (TEMP.
004781 (TEMP.
0262781 TEMP.
026 155 (TEMP.
2)
2)
2)
2)
Nissan 1980 49-state
* 0.
+ 0,
+ 0
•*• o.
,004667 (TEMP.
,004400 (TEMP.
.00739HTEMP.
,OO6466(TEMP.
2)
2)
2)
2)
Nissan 1980 Califorrv
* 0
f 0
.007627 (TEMP.
.007219(TEMP.
-------
3-3
Models Predicting Cold-Start CO Emissions for Vehicle 8R10Y131366. Ford 1978 California PINTO (140 cid)
N iemp inierva i equation K-5QK St
3
8 '
8
8
6
€
6
6
Models
N
7
7'
7
7
5
5
5
5
Models
N
12
12
12
12
10
10
10
10
20.0 - 100.0
20.0 - 100.0
20.0 - 100.0
20.0 - 100.0
20.0 - 75.0
20.0 - 75.0
20.0 - 75.0
20.0 - 75.0
Predicting Co
Temp Interval
20.0 - 10O.O
20.0 - 10O.O
20.0 - 100.0
20.0 - 1OO.O
2O. 0
20.0
20. 0
20.0
Pree
Temp
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
- 75.0
- 75.0
- 75.0
- 75.0
licting Co
Interval
- 1OO.O
- 1OO.O
- 1OO.O
- 100.0
- 75.0
- 75.0
- 75.0
- 75.0
Models Predicting Co
N Temp Interval
9
9
9
9
6
8
5
8
20. 0
20.0
20.0
20.0
20.0
20.0
2C.O
*1/^. '"'
- 10O.O
- 100.0
- 1OO.O
- 100.0
- 75.0
- 75.0
- 75.0
- 75.0
BAG1 CO = 108 . 14
BAG1 CC =• 137.33
(8AG1 - BAG3) CO = 87.23
(BAG1 - BAG3) CO = 99.72
BAG1 CO = 118.69
BAG1 CO =150.96
(BAG1 - BAG3 ) CO » 91 .49
(BAG1 - BAG3) CO = 108.69
Id-Start CO Emissions for
3AG1
BAG1
(BAG1
(BAG1
BAG1
SAG1
(BAG1
(BAG1
CO
CO
- BAG3)
- BAG3)
CO
CO •
- BAGS)
- BAG3)
CO
CO
CO
CO
3
3
3
S7.77
38.62
64.55
82.66
= 75.70
» 102.41
= 71.37
= 95.53
Id-Start CO Emissions
3AG1
BAG1
(BAG1
(BAG1
BAG1
BAG1
(BAG1
(BAG1
CO
CO
- BAG3)
- BAG3)
CO
CO
- BAG3)
- SAG3)
CO
CO
CO
CO
3
S
3
5
a
3
Id-Start CO Emissions
BAG1
BAG1
(BAG1
(BAG1
BAG1
5AG1
(SAG1
(3AG1
CO
CO
- BAG3)
- BAG3)
CO
CO
- BAG3 1
- 3AG3 I
CO
CO
CO
CO
S
3
3
S
S
=
=
for
16.41
22.75
15.40
21 .63
18.90
20.83
17 .81
2O.21
for
56. 10
79.74
53.21
77.57
66 . 29
79.59
63 . 78
75 . 54
- 1 . 1267(TEMP)
- 2.4607(TEMP) + 0.01 1384(TEMP.2)
- 0.8985(TEMP)
- 1.4695(TEMP) + 0.004872 (TEMP. 2)
- 1.3872(TEMP)
- 3. 3 125 (TEMP) + 0 .020948 ( TEMP . 2 )
- 1 .0037 (TEMP )
- 2.0305(TEMP) + 0 .01 1 172( TEMP . 2 )
Vehicle 4M47AAH202725 , GM 1980 50-S
- 0.6342(TEMP
- 1.6013(TEMP
- 0.6899(TEMP
- 1.5298(TEMP
- 0.8469(TEMP
- 2.4794(TEMP
- 0.8728(TEMP
- 2.3490(TEMP
Vehicle FB289,
- 0. 15O3(TEMP
- 0. 457 1( TEMP
- 0. 1469(TEMP
- 0.4487 (TEMP
- 0.2172(TEMP
- 0.3348 (TEMP
- 0.2117(TEMP
- 0.3581 (TEMP
Vehicle D162,
- 0.5492(TEMP
- 1.6038 (TEMP
- 0.5412(TEMP
- 1.6277(TEMP
- 0.8007 (TEMP
- 1 . 5939( TEMP
- 0.30201 TEMP
- 1.5635(TEMP
)
) + 0.008 171 (TEMP .2)
)
) + O.OO7096(TEMP.2)
)
) + 0.018352(TEMP.2)
)
) * 0.016595(TEMP.2)
Nissan 1981.5 49-Sta
)
) + 0.002740(TEMP.2)
)
) + 0.002696 (TEMP. 2)
)
) •*• 0.001322(TEMP.2)
)
) + 0.001646(TEMP.2)
Plymouth 1981 49-Stat
)
) + O.OO8742(TEMP.2)
)
) * 0.009006 (TEMP. 2)
)
) - 0.003631 ( TEMP. 2)
)
) * O.C08236I TEMP . 2 )
92 . 636%
98.843%
95.672%
97.519%
96. 128%
99. 173%
95.020%
96 . 656%
itate REGA
R-SQR
88.452%
97.336%
9 1 . 975%
97 . 883%
92 . 307%
97 . 980%
93 . 724%
98 . 1 58%
te 280ZX
R-SQR
65 . 35O%
80 . 695%
65.076%
79 . 344%
79 . 359%
79 . 744%
77 . 4O4%
78.016%
e RELIANT
R-SQR
88 . 577%
99 . 1 94%
84 . 323%
98 . 989%
97.392%
98 . 964%
97 . 380%
98 .324%
10.645
4.6213
S.4039
5.3116
7 .9148
4.2231
6.5325
6. 1817
i. (231 dd)
SE
8.3322
4 .4691
7 . 3988
4.2490
7. 1699
4 . 5000
6.6236
4.3943
TURBO (168 Cid)
SE
3.3588
2.6817
3.376O
2.7368
2.8127
2.9788
2 . 9O5 1
3.0633
(135 Cid)
SE
7.5771
1 .9352
7.8418
2. 1513
3.7251
2.7112
3.7397
2. 8933
-------
3-4
Models Predicting Cold-Start CO Emissions for Vehicle MA46100183,
Toyota 1980 50-State CELICA SUPRA (156 cid)
N Tano interval equation K-SUK st
14
14
14
14
12
12
12
12
Models
N
4
4
4
4
3
3
3
3
Models
N
4
4
4
4
3
3
3
3
Models
N
4
4
4
4
3
3
3
3
20.0 - 100.0
20.0 - 100.0
20.0 - 100.0
20.0 - 10O.O
20.0 - 75.0
20.0 - 75.0
20.0 - 75.0
20.0 - 75.0
Predicting Co
Temp Interval
22.0 - 97.0
22.0 - 97.0
22.0 - 97.0
22.0 - 97.0
22.0 - 72.5
22.0 - 72.5
22.0 - 72.5
22.0 - 72.5
i Predicting Co
Tamp Interval
23.0 - 98.0
23.0 - 98.0
23.0 - 98.0
23.0 - 98.0
23.0 - 75.0
23.0 - 75.0
23.0 - 75. O
23.0 - 75.0
; Predicting Co
Temp Interval
22.2 - 10O.O
22.2 - 1OO.O
22.2 - 10O.O
22.2 - 100.0
22.2 - 74.0
22.2 - 74.0
22.2 - 74.0
22.2 - 74.0
BAG1
8AG1
(BAG1
(BAG1
8AG1
8AG1
(8AG1
(8AG1
CO
CO
- BAGS)
- BAG3)
CO
CO
- BAG3)
- BAGS)
S
3
CO =
CO =
a
B
CO »
CO =•
41 .60
54.46
41 .09
53.65
46.53
53.56
45.91
52.62
- 0.4448 (TEMP)
- 1 .0045(TEMP)
- 0.4424(TEMP)
- 0.9889 (TEMP)
- 0.5392(TEMP)
- 0.9494 (TEMP)
- 0.5347(TEMP)
- 0.9259(TEMP)
+ 0.
+ 0.
* 0.
+ 0.
004906(TEMP
004790(TEMP
0043 141 TEMP
0041 13(TEMP
.2)
.2)
.2)
.2)
Id-Start CO Emissions for Vehicle B3BK26B4BC184143, Chrysler
BAG1
BAG1
(BAG1
(BAG1
BAG1
BAG1
(8AG1
(BAG1
CO
CO
- BAGS)
- BAGS)
CO
CO
- BAGS)
- BAGS)
S
a
CO *
CO =
s
3
CO =
CO =
142.56
216.85
135.25
209 . 36
168.98
254 .28
161 .27
251 .63
i Id- Start CO Emissions for
BAG1
BAG1
(BAG1
(BAG1
BAG1
BAG1
(BAG1
(BAG1
CO
CO
- BAGS)
- BAGS)
CO
CO
- SAGS)
- BAGS)
*
*
CO *
CO »
•
a
CO •
CO -
87.48
144.79
70.74
113.56
1O6.30
162.27
34.87
125.83
ild-Start CO Emissions for
BAG1
BAG1
(SAG1
(BAG1
5AG1
BAG1
(BAG1
( BAG1
CO
CO
- BAGS)
- BAG3)
CO
CO
- 3AG3)
- BAG3 )
3
3
CO *
CO =
s
CO *
CO =
- 1.5394 (TEMP)
- 4.8156(TEMP)
- 1.4874(TEMP)
- 4.7563(TEMP)
- 2. 18OKTEMP)
- 7.02O6(TEMP)
- 2. 1188(TEMP)
- 7.2458(TEMP)
* 0.
* 0,
+ 0.
* 0,
,028103(TEMP
.028040(TEMP
,052822(TEMP
,055948(TEMP
Vehicle 1 FABPO82XBW20356 . Ford
b\4fl V 1 Wt 1
- 0.9255(TEMP)
- 3.3876(TEMP)
- 0.7947(TEMP)
- 2. 6341 (TEMP)
- 1 .3747(TEMP)
- 4. 3683 (TEMP)
- 1.1320(TEMP)
- 3.3226(TEMP)
* 0
* 0
•*• 0
* 0
.0207 16 (TEMP
.015476CTEMP
.031388(TEMP
.022969(TEMP
Vehicle 1FABP21B3BK15310, Ford
147. 6O - 1 .7355(TEMP)
262. 10 - 6.6433(TEMP)
142.40 - 1 -6900(TEMP)
251 .74 - 6.3769(TEMP)
190.04
307 .74
183 .01
294 . 35
- 2.7480(TEMP)
- 9 . 23701 TEMP )
- 2.S587ITEMP)
- 3.8445I7EMP)
+ 0
+ 0
•*• c
* \J
.040848 (TEMP
.039009 (TEMP
.0700291 TEMP
.0662451 TEMP
.2)
.2)
.2)
.2)
1981
.2)
.2)
.2)
.2)
1981
.2)
.2)
.2)
.2)
89 . 463%
98 . 207%
89.813%
98 . 273%
97.287%
98.028%.
97.416%
98. 101%
1981 K-CAR
R-SQR
32.481%
98.769%
81 .237%
98 . 345%
93 . 755%
100.00%
92 . 668%
100.00%
3. 7156
1 . 6008
3 . 6264
1 .5594
1 .9615
1 .7629
1 .3974
1 .7144
(135 cid)
SE
27.213
10.201
27.419
11 .516
20 . SOS
0.000
22.036
O.OOO
ESCORT WAGON (98 Cid)
R-SQR SE
77. 191%
99 . 1 36%
8 1 . 79O%
99.388%
92 . 520%
100.00%
93 . 998%
100.0O%
FAIRMONT
R-SQR
72 . 709%
98 . 626%
73.521%
98 . 724%
91 .939%
1 00 . 00%
32.267=4
1 00 . 00%
19 . 444
5 . 3520
14 . 494
3.7578
14.689
0.000
10.749
O.OOO
(200 Cid)
SE
42.317
13.427
40.366
12.533
30.670
0 . 000
29.013
0.000
-------
3-5
Models Predicting Cold-Start CO Emissions for Vehicle 1G1AX68X7B621755. GM 1981 CITATION (173 cid)
r»
4
4
4
4
3
3
3-
3
Models
N
3
3
3
3
Models
N
6
6
6
6
4
4
4
4
Models
N
17
17
17
17
Models
N
17
17
' 7
i 7
iwup
22.0
22.0
22.0
22.0
22.0
22.0
22.0
22.0
Pree
Temp
18.7
18.7
18.7
18.7
Pree
Temp
20.0
20. 0
20.0
20.0
20.0
20.0
20.0
20.0
Pree
Temp
20.0
20.0
20.0
20.0
; Pret
Temp
20.0
20.0
20.0
20. C
interva i
- 1O1 .5
- 101.5
- 101 .5
- 101.5
- 75.0
- 75.0
- 75.0
- 75.0
iicting Co
Interval
- 73.6
- 73.6
- 73.6
- 73.6
licting Co
Interval
- 1OO.O
- 1OO.O
- 10O.O
- 1OO.O
- 75.0
- 75.0
- 75.0
- 75.0
licting Co
Interval
- 75.0
- 75.0
- 75.0
- 75.0
licting Co
Interval
- 75.0
- 75.0
- 75.0
- 75.0
BAG1
BAG1
(BAG1
(BAG1
BAG1
SAG1
(BAG1
(BAG1
CO
CO
- 3AG3)
- BAG3)
CO
CO
- BAG3)
- BAG3)
a
CO =
CO =
S
S
CO =
CO =
66.61
87.55
63.76
84.49
74 .87
91 .01
72. 12
85.28
Id-Start CO Emissions for
BAG1
BAG1
(BAG1
(BAG1
CO
CO
- SAG3)
- BAG3)
a
CO =
CO =
173.81
251 . 10
171 .82
253.35
Id- Start CO Emissions for
BAG1
BAG1
(BAG1
(BAG1
BAG1
BAG1
(BAG1
(BAG1
CO
CO
- BAG3)
- BAG3)
CO
CO
- BAG3)
- BAG3)
a
s
CO «
CO =
3
a
CO =
CO =
104.27
154.28
1O1 .55
152.23
125.57
164.20
123. 15
162. 10
Id-Start CO Emissions for
3AG1
BAG1
(BAG1
(BAG1
CO
CO
- BAG3)
- BAG3)
s
a
CO =
CO =
203 . 48
2O8 . 03
196.30
201 .20
ild-Start CO Emissions for
SAG1
BAG1
(BAG1
(3AG1
CO
CO
- 3AG3)
- 3AG3 )
=
CO =
CO =
33.47
63.05
33 .03
55 .31
equation
- 0.6064 (TEMP)
- 1 .5036 (TEMP)
- 0.6142(TEMP)
- 1 .5026 (TEMP)
- 0.8017(TEMP)
- 1.7054CTEMP)
- 0.81 19(TEMP)
- 1.5488 (TEMP)
* 0
* 0
+ 0,
* 0.
.OO74Q4(TEMP.2)
.007332(TEMP.2)
,009612(TEMP.2)
007837 (TEMP. 2)
Vehicle 2G1AW89U6B 144479, GM 1981
- 2.4327(TEMP)
- 7.3439(TEMP)
- 2.4367(TEMP)
- 7.6276(TEMP)
+ 0.
+ 0.
,055104 (TEMP. 2)
058399(TEMP.2)
Vehicle D160, Chrysler 1981 K-CAR
- 1.0856(TEMP)
- 3.2073(TEMP)
- 1.0821 (TEMP)
- 3.2326(TEMP)
- 1.5527(TEMP)
- 3.818O(TEMP)
- 1.5559(TEMP)
- 3.8399(TEMP)
* 0.
+ 0.
«• 0.
+ 0.
Vehicle 1009, Ethyl
- 2.4633(TEMP)
- 2.6914(TEMP)
- 2.4769(TEMP)
- 2.7219(TEMP)
* 0.
+ 0,
.01737KTEMP .2)
0176O6(TEMP.2)
024032(TEMP.2)
,024232(TEMP.2)
Lean -Burn Car
,OO2318(TEMP.2)
,00249O(TEMP.2)
K-5QK
90 . 8 1 3%
99.922%
91 .245%
99 . 996%
98. 171%
100.00%
98 . 8O7%
100.00%
MALZBU (267
R-SQR
94 . 264%
100.00%
93 . 394%
100.00%
SE
7.8367
1.0192
7.7294
0.2334
4.2385
0.000
3.4562
O.OOO
cid)
SE
24.263
O.OOO
25 . 124
O.OOO
(WON) (135 Cid)
R-SQR SE
84 . 355%
99 . 785%
83.913%
99 . 784%
96 . 9 1 3%
99 . 996%
96.871%
99.991%
(163 Cid)
R-SQR
96 . 1 58%
96 . 1 75%
95 . 89O%
95 . 909%
Vehicle 1X89L7U 104103, Dresser 1977 NOVA (350
Equation • R-SOR
- 0.4513(TEMP)
- 1.8657 (TEMP)
- 0.471 1 (TEMP )
- 2. 01 45 I TEMP)
+ 0.
* 0.
0143501 TEMP . 2)
0156591 TEMP . 2 )
74.603%
93.274%
74 . 3O6%
94 .889%
15.564
2. 1075
15.773
2. 1128
8.8177
0.4561
8.8977
0.676O
SE
10.9O5
11 .263
11.357
11 .728
cid)
SE
5.2759
2.8104
5 . =207
2 . 5586
-------
APPENDIX 4
Correlation Coefficients between:
1. Difference in (actual) CO emissions from Bag 3 to Bag
1 and the corresponding differences in fuel
consumption. (Stratified by temperature)
2. Actual CO emissions (from Appendix 1) and vehicle
parameters.
3. Predicted CO emissions (from Appendix 3) and vehicle
parameters.
4. Pairs of vehicle parameters.
KEY;
AIR: 0 = No secondary air; 1 = Has secondary air.
PMP/PLS: 0 « Pulse air; 1 = Air pump. (Missing - No air.)
GXD: 1 = Has only oxidation catalyst; 0 = Otherwise.
3WY: 1 = Has only three-way catalyst; 0 = Otherwise.
BOTH: 1 = Has both oxidation and three-way catalysts;
0 = Otherwise.
#CAT: Number of catalyst(s) (0, 1, or 2)
TYPE: 0 = Monolith Catalyst; 1 = Pelleted Catalyst.
FS : 0 = EFI; 1 = Carbureted. (FS: Fuel Metering System)
A/F: 0 = Open-loop air/fuel (A/F) control;
1 = Closed-loop A/F.
EGR: 0 = No Exhaust Gas Recirculation (EGR); 1 = Has EGR.
ETW: Equivalent Test Weight (pounds)
ICYL: Number of cylinders (4, 6, or 8)
4-1
-------
4-2
Correlation Matrices for ACTUAL Emission & Fuel Consumption Data
(from Appendix 1).
Correlation Coefficients between the Increase in Bag 3 to Bag 1 CO
Emissions and the Corresponding Increase in Fuel Consumed for All
18 Vehicles.
Correlation
Temp Coefficients
20°F 0.7159 (N» 33 DF= 31 R@ .0500= .3440 R@ .0100= .4421)*
30°F N/A (TOO FEW CASES FOR ANALYSIS)
40°F 0.9638 (N= 7 DF= 5 R@ .0500= .7545 R@ .0100= .8745)
60°F 0.7799 (N= 36 DF= 34 R@ .0500= .3291 R@ .0100= .4238)
75°F 0.4721 (N= 50 DF= 48 R@ .0500= .2787 R@ .0100= .3610)
100°F -0.1991 (N= 26 DF= 24 R@ .0500= .3882 R<5> .0100= .4958)
Correlation Coefficients between the Increase in Bag 3 to Bag 1 CO
Emissions and the Corresponding Increase in Fuel Consumed for 16
Vehicles (All Cars Except the Ethyl and Dresser Cars)
Correlation
Temp Coefficients
20°F 0.6972 (N= 27 DF= 25 R@ .0500= .3809 R@ .0100= .4869)
60°F 0.8038 (N= 28 DF= 26 R@ .0500= .3739 R@ .0100= .4785)
75°F 0.5006 (N= 38 DF= 36 R@ .0500= .3202 R@ .0.100= .4128)
100°F -0.1991 (N= 26 DF= 24 R@ ..0500= .3882 R@ .0100= .4958)
"N" is sample size, 'DF' is degrees of freedom.
'R@ .0500= .3440' indicates that the correlation coefficient which
corresponds to the '.0500' significance level is '.3440'
-------
4-3
Correlation Coefficients between Pairs of Actual Data (Appendix 1) and Vehicle Parameters,
Stratified by Ambient Temperature:
Correlation Matrices for All Vehicles Except the Ethyl & Dresser Cars
ACTUAL
VARIABLE
Temperature
Bag 1 CO
*
Delta Fuel
Delta CO
Temperature
Sag 1 CO
Delta Fuel
Delta CO
Temperature
Bag 1 CO
Delta Fuel
Delta CO
= 20'F, N = 27 DF= 25
.6330 .7436 .0554
.3372 .4151 -.0696
.6096 .7152 .0620
= 60'F, N= 28 OF* 26
.4024 .4687 .4385
.2372 .2798 .3220
.3869 .4509 .4265
= 75"F, N= 38 DF = 36
.4007 .4649 .4991
.1069 .1393 .3675
.3525 .4166 .4938
AIR PMP/PLS 0X0
R .0500= .3809
-.5329- .4979
-. 1787 .2227
-.5014 .4622
R .0100= .4421 :
-.0139 -.1472
-.1186 .2259
-.03O1 -.1184
R@ .0100= .4238:
- . 1464 -. 1595
- . 1614 . IOCS
-. 1Q44 -. 1220
R® .0100= .3610:
-.2782 -.3198
-.1810 .0050
-. 1533 -.2177
#CAT TYPE
.7570 -.1727 .2346 .4438 .3417
.4257 .2251 .0649 .6916 .6149
.7275 -.1551 .2334 .4925 .4087
4808
2426
4604
4895
1236
4395
FS
6561
4562
6373
3918
2359
363-i
3860
1276
2877
FS
- .2656
-.2193
-.2498
-.3946
-.2623
- . 3635
A/F
-. 1961
- . 0308
- . 1917
- . 1412
? Q 1 ^
-.103'.
-.2635
- . 2198
-. 1545
A/F
. 1629
. 1223
. 1853'
.0468
.0818
.0625
EOT
.2206
. 1069
.2217
. 1342
• n =; .-^,
. 1472
.0422
.0852
.0303
EGR
.6484
.3043
.6451
.5275
.6042
.5025
ETW
,
J4J104
. 1007
.2423
.4659
. 2-73
. 1539
.3983
. 1809
ETW
. 5392
. 7459
.5547
-.2826
.5323
.2906
. 1497
.5143
.2082
. 2T7 1
.5502
.2356
.0307
.4229
.0880
*CYL
Delta Fuel = Total fuel consumed (in grams) during Bag 1 minus total fuel consumed during
the corresponding Bag 3.
Delta CO = CO Emissions (grams per mile) during Bag 1 minus corresponding Bag 3 CO Emissions.
-------
Correlation Matrices for PREDICTED Bag 1 CO (Appendix 3) for All 18 Vehicles
Correlation Coefficients for Linear Model with Ho Data from Tests Over 90'F
N» 17 DF = 15' R@> .0500= .4821 R* .0100= .5055
PREDICTED
VARIABLE
CC » 20' F
CO 3> 40' F
CO 0 75' F
Slope»20' F
S 1 ope«»40' F
Slope@75' F
RATIO*
Sen% 20' F
Sen » 40' F
Sen > 75' F
DIFF
Correlation
N= 18 DF=
PREDICTED
VARIABLE
CO 9 20' F:
CO » 40' F:
CO * 75' F:
Slope@20' F
S 1 opee40' F •
S 1 ope@75' F
RATIO
Sen * 20' F
Sen a 40' F
Sen 9 75' F
DIFF
RATIO «
OIFF =
?on ( 'Zor
.5981
.5843
. 4 174
- . 5097
-.6097
-.5097
.0663
-.0829
-.0739
- .0663
.6097
AIR
Coeff
16 Re
.5867
.5946
.5300
-.5379
-.5379
- . 5379
.0024
. 1054
. 1079
- .0024
.5379
AIR
Bag 1
Bag 1
\« i -f i w i
.7155
.6925
.46OO
-.7434 -
-.7434 -
-.7434
. 1392 -
- . 2707
-.2522
-. 1392
.7434
PMP/PLS
icients for
.2035 -.5435
.2536 -.5255
.4792 -.3462
.0864 .5658
.0864 .5658
.0864' .5658
.1919 - . 1460
.5049 . 1332
.4783 .1350
.1919 . 1460
.0864 -.5658
0X0 3WY
.2351
. 1897
- . 08 1 3
-.3277
- . 3277
- .3277
.2926
-.4265
-.4178
-.2926
.3277
BOTH
Linear Model with Data
.0500= .4683 R9 .0100= .
.7038
.7035
. 5835
-.6685 -
- . 6685 -
-.6635 -
. 0603 -
-.0952
-.0702
- . 0603
.6685
PMP/PLS
CO at 20' F
CO at 20' F
t\/ > = '
.1812 - .5415
.2250 -.5364
.3833 -.4236
.069O .5255
.0690 .5255
.0690 .5255
. 1450 - .0815
.^493 -.0040
.3993 ..0005
. 1450 .0815
.0690 -.5255
OXD 3WY
divided by Bag
minus dy Bag 1
'_ d(Tsmp)
.5897
.2351
.2195
.1130
-.2595
-.2595
- .2595
. 1730
- . 1696
- . 1492
-. 1729
.2595
BOTH
1 CO at
CO at 75
-.0197
-.0510
- . 2089
-.050O
- . 0500
- . 0500
.2297
-.3057
- . 2996
-.2297
.0500
/CCAT
-.0467
-.0686
-. 1738
- . 0030
- . 0030
- . 0030
.4146
-.3199
- .3344
- . 4146
.0030
TYPE
from Tests up
-.0256
-.0249
-.0176
.0260
.0260
.0260
. 1219
-.0555
-.0375
- . 1218
- .0260
75' F. '
• c
• -.0642
-.0771
-. 1215
.0308
.0308
.0308
.4160
- . 254 1
- . 2905
- .4150
- .0308
TYPE
.6083
.5848
.3642
- .6423
- .5423
-.5423
.2561
-.5296
- .51 18
-.2561
.6423
FS
to 105 'F
.6054
.5978
.4636
-.5919
-.5919
- . 5919
. 1986
- .-374
- .4055
- . 1986
.5919
FS
-. 1580
- . 1778
- . 2540
. 1095
. 1095
. 1095
.0455
.0022
.0183
-.0455
-. 1095
A/F
-.2111
- . 207 1
- . 1547
.2094
.2O94
.2094
.0342
. 1326
. 1536
- .034'1
- . 2094
A/F
.2736
.2527
. 1014
- .31 10
-.3110
- .31 10
. 1780
-.3964
-.3825
-. 178O
.3110
EGR
.270O
.2557
. 1485
- . 2898
- .2898
- . 2898
. 1275
- . 1950
- .2001
-. 1276
.2898
EGR
.2833
. 3140
.4269
- . 2065
- . 2065
- . 2065
.1721
.0887
.0664
- . 1721
.2065
ETW
.3060
.3166
.3106
- .2654
-.2654
- . 2654
.2545
- . 1575
- . 1959
- .2546
.2654
ETW
.2491
• .2586
.2679
- . 2202
- . 2202
-.2202
. 3109
-.0153
- . 04O4
- . 3 1 09
.2202
,CYL
.28O4
.2579
. 1 145
- .3184
-.3"! 84
- . 3184
.41 15
- . 3333
- . 3651
-.4115
.3184
#CYL
CO
Temp
-------
4-5
Correlation Matrices for PREDICTED Bag 1 CO (Appendix 3) for All 18 Vehicles
Correlation Coefficients for Quadratic Model with No Data from Tests Over 90'F
N = 18 OF = 16 R@ .0500= .4683 R@ .0100= .5897
PREDICTED
VARIABLE
CO ® 20' r
CO <& 40' F
CO * 75' F
SI 006920' F
SI 006040' F
S10De*75' F
RATIO
Sen -3 20' F
Sen * 40' F
Sen 9 75' F
DIFF
Corre 1 at i on
N = 17 OF=
PREDICTED
VARIABLE
CO 9 20' F
CO » 40' F
CO » 75' F
S1opa»20"F
Slope®40' F
'Slooe@75' F
RiTIO
Sen '9> CO' F
Sen s. JQ' "
Sen a 75' F
OIFF
.8143
.4808
.3860
- . 421 1
-.5912
-.2413
. 1305
.06O6
- . 0086
.0708
.6076
AIR
Coeff
15 R@
.5009
.5470
.4180
-.5454
- . 609 1
- . 4J72
.0398
.2788
. 1773
- . 4098
.6160
AIR
.7343
.5691
.4236
-.5101
- .7187
-.2968 -
.1711 -
.0646
- . 0664
.0859 -
. 7396
PMP/PLS
icients for
. 1506
.3106
. 4741
.2321
.0706
.3030
. 1085
.2250
.2554
.0824
.0166
0X0
-.5756
- . 4154
- . 2905
.4554
.5936
. 1830
-. 1823
.0483
. 1409
- . 1421
-.5942
3WY
.3105
.0551
-. 1231
- . 546O
-.4981
. 1456
.2689
-.2755
-.3684
.2480
.4181
BOTH
Quadratic Model with
.0500= .4821 R»
.7207
.6436
.4662
-.6785
-.7472 -
- . 5259 -
. 1405 -
.2237
.0916
- .4531
.7522
PMP/PLS
. 1910
.2905
.4693
.0543
.0345
.2328
.2730
.2705
. 2361
.0669
.0669
OXD
.0100=
- . 55O9
-.4858
- . 3496
.5325
.5769
. 3851
-. 147S
- . 1319
- .0439
.2943
- . 5775
3WY
.6055
. 250O
. 1034
-.0643
-.4897
-.3943
.0429
.3472
- . 1374
- .2194
- . 1477
.3476
BOTH
.0523
-.1381
-.2391
-.3570
-.2336
.2525
.2224
-.2673
-.3216
.2261
. 1466
#CAT
Data from
-.0108
- . 1344
- . 1933
-.2629
-. 1 178
.2965
.2717
-.2320
- .2035
- .0127
.06O1
0CAT
-.0798
-.0499
-. 1088
.0866
.0733
- .0328
- . 0670
- . 0503
-. 1690
-.0582
-.0585
TYPE
Tests
-.0475
-. 1463
-.1137
-. 1947
- . 0409
.3546
. 3742
-.4381
- .4436
. 549O
- .0179
TYPE
.6386
.4804
.3480
-.4697
-.6395
-.2351
. 1569
-.0652
- .2514
. 1047
.6503
FS
up to 1O5
.6272
. 541 1
.3791
- .6322
- .6694
-.4119
. 3470
- . 0609
-.2354 •
-. 1659
.3648
FS
-.2495
-. 1892
-. 1303
. 1799
.250O
.0987
- .3477
.2017
.3114
- .3645
-.25SO
A/F
"f
•
- . 1721
-.2279
- .2383
.0024
.0979
.2795
- .0804
.0559
. 1 175
- .0486
- . 1310
A/F
.2952
.2349
.0689
-.1811
- . 3O6 1
- . 1923
. 1077
.0615
-. 1310
.0261
.3327
EG*
.2818
.2348
. 1 145
-.2948
-.'3197
-.2139
.2460
- .O546
- . 1 .7 1 2
-. 1303
.3201
EQR
.2577 .2077
.4O86 .3576
.4166 .2675
.2016 .2265
-.0488 -.0338
-.4512 -.4710
.0673 .0191
.2725 .3265
.0624 .084O
.0367 -.0494
. 1661 . 1576
ETW #CYL
.2683 .2325
.3102 .2393
.4784 .3264
-.1320 -.1716
-.1579 -.1779
-.1388 -.1006
.0442 . 1544
-.0250 -.0915
- .0945 - . 1800
.4847 .5017
. 1632 . 1753
ETW #CYL
-------
4-6
Correlation Matrices for PREDICTED Bag 1 CO (Appendix 3) for All Vehicles Except the Ethyl & Dresser Cars
Correlation Coefficients for Linear Model with No Data from Tests Over 90'F
N= 15 DF:= 13 R» .0500s .5140 R0 .0100= .5411
PREDICTED
VARIABLE
CO » 20' F:
CO * 40 l:
CO e 75' F:
S 1 ope 75' -
S1oqe*20' F
Slope@40' F
Slope*75" F
RATIO
Sen ® 20' F
Sen * JO' F
Sen $ 75' F
DIFF
.5454
.5260
.3325
-.5676
-.5676
-.5676
. 1886
- . 305O
- . 304S
-. 1887
.5676
AIR
.6797
.6497
.3799
- . 7202 -
-.7202 -
-.7202 -
.2642 -
-.5106
-.4998
-.2642
.7202
PMP/PLS
Coefficients for
14 R®
.539O
.5391
.4522
- . 5084
- . 5084
- . 5084
.1301
-.1105
- . 1255
- . 1301
. 5084
AIR
.2311 -.4737
.2347 -.4515
.5080 -.2574
. 1041 .5046
. 1041 .5046
.1041 .5046
.1837 -.2460
.5071 .2946
. 48O5 . 3062
. 1837 .2460
. 1041 - .5047
0X0- 3WY
Linear Model
.0500" .4973 R* .0100"
.6733
.6643
.5184
-.6570 -
-.6570 -
-\6570 -
. 1859 -
- .3444
- .3302
-.1359
.5570
PMP/PLS
.2110 -.4765
.2523 -.4659
.3876 -.3458
.0998---' .4749
.0998 .4749
.0998 .4749
. 1288 - .1323
.4516 .1518
.4054 .1732
. 1288 . 1823
.0999 -.4749
OXD 3WY
.29O1
.2327
-.1030 -
-.4059 -
-.4059 -
-.4059 -
.3576
-.6239 -
-.6167 -
-.3576 -
.4059
BOTH
with Data
.6226
.3O45
.2672
.0647 -
-.3771 -
-.3771 -
-.3771 -
.2562
-.4495 -
- . 4321
-.2562 -
.3771
BOTH
. 1720
. 1 170
. 1852
.2868
.2868
.2868
.3215
.6180
.6041
.3215
.2368
#CAT
from
. 1883
. 1488
.0437
.3156
.2785
.0391
-.3854
-.3854
-.3854
.4252
- .3269
-.3323
-.4252
. 3854
TYPE
Tests up
.3058
.2646
.0463
.2726 -.3875
.2726
.2726
.2344
.4953
.4576
.2344
.2726
#CAT
- .3875
-.3875
.4469
-.3721
-.3933
- .4469
.3875
TYPE
.6747
.6488
.4002
- . 7O62
-.7062
-.7062
.2455
-.5158
- . 4986
-.2455
.7062
FS
to 105 'F
.6698
.6639
.5315
-.6462
- .6462
- .5462
. 1750
-.4117
-.3772
- . 175O
.6462
FS
-. 1826
-.2159
- .3473
. 1O24
. 1024
. 1024
. 1292
-.2391
-.2186
-. 1291
-. 1024
A/F
-.2303
-.2484
-.2854
. 1738
. 1738
. 1738
. 1409
- . 1743
- . 1S23
- . 1409
- . 1738
A/F
.2979
.2755
. 1 125
-.3356
-.3356
-.3356
. 1694
- . 388O
-.3747
- . 1694
.3356
EGR
.2929
.2799
. 1794
- . 3075
-.3075
- . 3075
. 1 106
- . 1563
-. 1633
- . 1 106
.3075
EGR
.5311
.5665
.6468
-.4326
-.4326
-.4326
. 1017
.2474
. 2308
- . 1017
.4326
ETV
.5659
.5717
.5041
- . 5200
- . 5200
- . 5200
. 1S32
- .0752
- . 099=
-. 1932
.5200
ETX
.3867
.3989
.3902
-.3450
-.3450
-.3450
.2657
. 1295
. 1059
-.2657
.345O
#CYL
.4229
.4OO7
.2414
-.4524
- .4524
-.4524
.3527
- . 2 193
- .2539
- .3627
.4524
#CYL.
-------
4-7
Correlation Matrices for PREDICTED Bag 1 CO (Appendix 3) for All Vehicles Except the Ethyl & Dresser Cars
Correlation Coefficients for Quadratic Model with No Data from Tests Over 90"F
N= 16 DF = 14 R@ .0500= .4973 R@ .0100= .6226 '
PREDICTED
VARIABLE
CO 9 20' F
CO * 40' F
CO » 75' F
S1ope®2O' F
S!ope<3>4O' F
Slooe 75' F
DIFF
Correlation
N = 15 DF=
PREDICTED
VARIABLE
CO * 20' F
CO » 40' F
CO * 75' F
Slope4O' F
Slope®75' F
RATIO
Sen .a 20' F
Sen s. 4C" .r
Sen * 75' F
DIFF
.5708
.4171
.3100
-.4116
-.5592
-. 1732
. 14 19
- . 1298
- . 2471
. 1258
.5718
AIR
. 7074
.5156
.3508
- . 5086
- . 7009
- . 23O8 -
. 1888 -
- . 1 176
- . 3054
. 1443
.72O4
PMP/PLS
Coefficients for
13 R@
.5572
.4867
.341 1
-.5364
-.5802
-.3764
.2520
-.0215
- . ' 623
-. 1683
.5848
AIR
. 1684
.3452
.5007
.2412
.0732
.3336
. 1 178
.2309
.2600
.0913
.0229
OXD
-.5176
- .3364
- . 2054
.4417
.5517
. 1030
-. 198O
.218O
.3498
- . 1981
- .5454
3WY
.3714
.0842
- . 1419
-.5735
-.5643
.1261
. 2635
- .3576
- . 5002
.2459
.4944
BOTH
Quadratic Model with
.0500= .5140 RS>
.6925
.5959
.3947
-.6819
-.7340 -
- . 4660 -
.3570 -
-.0877
-.2517
- . 24O9
.7384
PMP/PLS
.2183
.3382
.4936
.0588
.0448
.3293
.2689
.3185
.3133
. 1 185
.0856
OXD
.0100=
-.4878
-.4055
-.2681
.51O3
.5286
.2781
- .3207
. 1539
. 2544
.0610
- . 5241
3WY
. 641 1
.3122
. 1527
-.0830
-.5223
-.469O
- .0384
.4862
- . 3624
- .4538
- . 1385
.4369
BOTH
.
.2509
- . OO59
- . 2208
-.4889
-.4505
. 1597
.2339
- . 32O6
- .4519
.2068
.3786
#CAT
Data from
. 1915
.0475'
-. 1705
-.4034
-.3465
.0244
. 4553
-.3186
- . -217
- . 1976
.3157
#CAT
.2351
.2727
..1 1 1 1
.0182
- . 1649
- . 3165
-. 1208
. 2243
.0631
- . 1971
. 24 14
TYPE
Tests
.3032
.2067
. 1O29
- . 4077
- .3734
-.0550
. 3224
-. 1562
-.-710
.2837
.3511
TYPE
.6997
.5186
.3740
- . 49O4
-.6818
-.2326
. 1727
-.0511
- .2466
. 1 19O
.7030
FS
up to 105
.6875
.6OO1
.4131
-.6613
-.7179
-.4728
. 3307
-.0234
- . 2210
- . 27 1 1
.7245
FS
-.2953
- . 2C50
-. 1636
.2332
.2960
.0625
- .4401
. 1893
.2911
-.459O
- . 2946
A/F
•F
-. 1881
-.2530
-.3245
.0182
.O879
.2458
.0613
- . 1028
- . 1OS2
.0594
- . 1 140
A/F
.3190
. 2491
.0742
-. 1893
-.3225
- . 1913
. 1 170
.0803
-. 1223
.0339
. 3548
EGR
.3027
.2536
. 1255
— . 3056
- . 3366
-^2352
.234-.
- .0343
- . -.645
-.2051
.3415
EGR
.4821 .3300
.6518 .4918
.6163 .3755
.1638 .2014
- . 1928 - . 1 106
-.6410 -.5720
.0691 .0306
.4964 .4902
.2703 .2410
-.0112 -.O699
.3566 .2606
ETW #CYL
.5013 .3570
.5807 .3781
.6912 .4469
-.2283 -.2339
-.3367 -.2738
-.4673 -.2351
- . 1059 . .C52S
.3319' .1306
. 2C3S . C233
.3144 .4569
.3729 .2837
ETW XCYL
-------
4-8
Correlation Matrices for All 18 Vehicles
Correlation Coefficients between pairs of Vehicle Parameters:
N = 18 OF = 16 R@ .0500= .4683 R@> .0100= .5897
VARIABLE
AIR 1.0000
PMP/PLS .9286 1.0000
OXD .1890 .0236 1.0000
3WY -.7559 -.6614 -.2500 1.0000
BOTH .5345 .5345 -.3536 -.7071 1.0000
*CAT .2224 .2224 -.2942 -.3922 . 832r 1.OOOC i
TYPE -.0279 .0279 -.1474 .3686 0. .1735 1.0000
FS .6786 .7679 -.2362 -.4725 .5345 .2224 .2229 1.OOOO
A/F -.0945 -.0472 -.5000 .2500 .2357 .3922 .2949 -.0945 1.0000
EGR .2362 .1890 .1250 -.5000 .3536 .2942 -.1843 .2362 -.2500 1.OOOO
ETW -.1013 -.0716 .4207 .2589 -.4053 -.2937 .5944 .0874 -.2034 -.1294 1.0000
»CYL -.1793 -.1793 .2372 .3162 -.4472 -.3721 .5595 0. -.1581 0. .8596 1.0OOO
AIR PMP/PLS OXD 3WY BOTH #CAT TYPE FS A/F EGR ETW #CYL
Correlation Matrices for 16 Cars (All Vehicles Except the Ethyl & Dresser Cars)
Correlation Coefficients between pairs of Vehicle Parameters:
N» 16 DF= 14 Re .0500= .4973 R* .0100= .6226
VARIABLE
AIR
PMP/PLS
OXD
3WY
30~H
-CAT
TY = E
f $
A/F
EGR
ETW
,-CYL
1 .OOOO
.9115
. 1816
-.7125
. 5447
. 2894
.2774
.332-;
- .2774
.3026
.0946
- .0402
AIR
1.OOOO
0. 1.OOOO
-.5970 -.2548 1.0000
.5578 -.4286 -.7545
.3266 -.4880 -.=919
.3651 -.2'182 .2335
. 9 12S - . 2182 - . 544S
-.1826 -.5547 .3892
.2390 .1429 -.5606
.1245 .4752 .1287
-.0529 .2845 .2368
PMP/PLS OXD 3WY
I J
'. . OOOO
.3783 1.0000
- .0727 -.1491 i.0000
S547 .4-72 .3333 i.OOCO
.0727 .1491 .333.3 0. 1.0000
.4286 .4880 -.2182 .2182 -.2182 1.OOOC
-.4370 -.5O45 .5159 .0787 -.2011 -.1546 1.0000
-.4109 -.4102 .5276 -.0483 -.0483 -.0315 .3676 1.OOOO
BOTH #CAT TYPE FS A/F EGR ETW #CYL
-------
APPENDIX 5
Regression Analyses Modeling Cold-Start
CO Emissions
The results from forward and backward selection indicate that
relationships can be determined which have R-squared values
that exceed 0.7. The relationships are summarized below.
i
Results From Backward Selection
----- Value of the Coefficient of the Term -----
Dependent Variable Constant Temp FS ETW FS*TEMP FS*TEMP2 R-Sq.
Bag 1 CO -134.310 0.852 109.59 0,021 -1.800 0.006 0.78
Bag 1 CO-Bag 1 CO -97.936 0.855 102.01 0.010 -1.846 0.007 0.76
Results From Forward Selection
--- Value of the Coefficient of the Term —
Dependent Variable Constant Temp2 ETW FS*TEMP FS*ETW R-Sq.
Bag 1 CO 73.719 0.009 -0.041 -1.131 0.031 0.77
Bag 1 CO-Bag 1 CO 93.998 0.009 -0.047 -1.153 0.029 0.76
@75°
FS is a categorical variable. If a venicle has fuel injection, FS =
0. If a vehicle is carbureted, FS = 1.0.
5-1
-------
5-2
SELECT OPTIQNS=STEPWISE,BACKWARD VAR=03;2,20,7,12,17,71;72,74,81,82
MAXIM=100 LEVELS*.05,.10>
Selection of Regression of Bag 1 CO (BONE) Using All Vehicles Except
Ethyl 4 Dresser Cars (16 Cars)
ANALYSIS At STEP 0 (Initial Step) FOR 3.BONE N= 119 OUT OF 119
SOURCE
REGRESSION
ERROR
TOTAL
MULTIPLE R= .88141
VARIABLE
CONSTANT
2. TEMP
20. TEMP. 2
7 .PLS
12. FS
17.ETW
71 .PLS-TEMP
72 . PLS-TEMP .2
74.FS-PLS
81 . FS'TEMP
82 . FS'TEMP .2
ANALYSIS AT STEP 5
SOURCE
REGRESSION
ERROR
TOTAL
MULTIPLE R = .88036
VARIABLE
CONSTANT
2. TEMP
12. FS
17. ETW
81 . FS-TEMP
82. FS-TEMP .2
REMAINING
20. TEMP. 2
7. PLS
71 .PLS'TEMP
72.PLS-TEMP.2
74.FS*PLS
DF SUM OF SORS
10 . 13340 *6
108 38312.
118 .17172+6
R-SOR= .77689 SE=
PARTIAL COEFFICIENT
-196.08
.09395 1.9510
- .05871 - . 10570 -1
.OOOOO 43.252
. OOOOO 164.36
.497O5 .20961 -1
- .01431 - . 17 138
.02097 .20800 -2
-.COOOO -44.638
-.31725 -2.3125
. 18052 . 1 1077 -1
MEAN SQUARE
13340.
354.74
18.835
STD ERROR
.61971 +8
1 .9894
. 17294 - 1
.61971 +8
.61971 +8
.35212 -2
1 . 1523
.95443 -2
.61971 -t-8
.66518
.58077 -2
(Final Step) FOR 3. BONE N= 119
DF SUM OF SORS
5 . 133O9 +6
113 38632.
118 . 17172 *6
R-SOR* .77503 SE*
PARTIAL COEFFICIENT
-134.31
.33372 .85150
.72186 .109.59
.49366 .20792 -1
-.61885 -1.7998
.40289 .64208 -2
PARTIAL SIGNIF
-.07818 .4083
-.03940 .6772
-.03181 :7369
-.03738 .6930
-.03940 .6772
MEAN SQUARE
26617.
341 .88
18.490
STD ERROR
18.737
.22627
9.3839
.34457 -2
.21491
. 13722 -2
r _ 5 T i f
37 .606
T-STAT
-.31641 -5
.98070
- .61 1 17
.69794 -6
.26522 -5
5.9529
- . 14873
.21794
-.72030 -€
-3.4766
1 .9073
OUT OF 119
F-STAT
77.856
T-STAT
-7. 1682
3.7633
1 1 . 088
6.0341
-8.3747
4.6793
SIGNIF
.0000
SIGNIF
1 . OOOO
.3289
.5424
1 . 0000
1 . OOOO
.0000
.8820
.8279
1 . OOOO
.0007
.0591
SIGNIF
.0000
SIGNIF
.0000
.OO03
.OOOO
.OOOO
.OOOO
.OOOO
-
REGRESSION OF 3 . SONE USING BACKWARD SELECTION
STEP R-SQR STD ERROR - VAR VARIABLE
PARTIAL SIGNIF
^
1
2
3
4
5
. 77689
.77689
.77688
.77648
.77640
.77503
18 .835
18.748
18 .663
18.595
18.515
18.490
10
9
8
7
6
5
0
9
8
7
6
5
7. PLS
74.FS«PLS
72.PLS*TEMP.2
71 . PLS-TEMP
20. TEMP .2
IN
OUT
OUT
OUT
OUT
OUT
. OOOOO
- .00420
.04247
-.01894
- .07818
1 .0000
.9651
.6566
.8422
.4083
-------
5-3
SELECT OPTIONS=STEPWISE,FORWARD VAR=03;2.20,7,12,17,71-74,81-83,91,92,93
MAXIM=100 LEVELS*.05,.10>
Selection of Regression of Bag 1 CO (BONE) Using All Vehicles Except
Ethyl 4 Dresser Cars i16 Cars)
ANALYSIS AT STEP 1 (Initial Step) FOR 3.BONE N= 119 OUT OF 119
SOURCE OF SUM OF SORS MEAN SQUARE F-STAT SIGNIF
REGRESSION
ERROR
TOTAL
MULTIPLE 5= .S3O87
VARIABLE
CONSTANT
2. TEMP
REMAINING
20. TEMP. 2
7.PLS
12. FS
17.ETW
71 .PLS-TEMP
72.PLS-TEMP .2
73.PLS-ETW
74.FS-PLS
81 .FS"TEMP
82.FS-TEMP.2
83.FS-ETW
91 .FS»PLS«TEMP
92.FS*PLS'TEMP.2
93. FS'PLS'ETW
ANALYSIS AT STEP 6
SOURCE
REGRESSION
ERROR
TOTAL
MULTIPLE R= .87827
VARIABLE
CONSTANT
2O. TEMP. 2
17. ETW
81.FS-TEMP v
83.FS»ETW
REMAINING
2 . TEMP
7.PLS
12.FS
7 1 . PLS'TEMP
72. PLS-TEMP .2
73.PLS-ETW
74. FS'PLS
82.FS«TEMP .2
91 .FS»PLS*TEMP
92.FS«PLS'TEMP.2
93 . FS»PLS*ETW
1
1 17
1 18
R-SQR=
PARTIAL
-.S3O87
PARTIAL
.23864
-. 15201
68342.
. 10337 +6
. 17172 +6
.39799 SE*
68342. 77.350
883.55
29.724
COEFFICIENT STD ERROR T-STAT
84.892
-.87305
SIGNIF
.0093
. 1003
S.9503 12.214
.99268 -1 -8.7949
.0000
SIGNIF
.0000
.0000
.49859. .OOOO
. 40492
-.06789
.06538
.05522
.43871
. 29072
.26129
. 60202
.27256
.26594
.,55526
(Final
OF
4
1 14.
1 18
R-SQR=
PARTIAL
.45367
-.53754
- .69676
.770t9
PARTIAL
- .09398
. 0090 1
- .04090
-.02375
- .01029
.00901
.00542
. 15276
- . 00009
.07951
. 0090 1
REGRESSION OF 3 . BONE USING
STEP R-SOR STD
1 . 39799 29
2 .51618 23
3 .67576 22
4 .67446 21
5 .71210 20
5 .77135 18
ERROR
.724
.836
.003
.952
.734
.558
.0000
.4651
.4818
.5526
.0000
.0014
.0043
.0000
.0028
.0036
.0000
Step) FOR 3.
SUM OF SORS
. 13245 *Q
39262.
. 17172 *€
.77135 SE=
BONE N* 119 OUT OF 119
MEAN SQUARE F-STAT
33114. 96. 147
- 344.41
18.558
COEFFICIENT STD ERROR T-STAT
73.719
.85788 -2
- . 41 108 - 1
-1.1313
. 31 124 - 1
SIGNIF
.3178
.3238
. 5543
.801 1
.9131
.9238
.9542
. 1O31
.9992
.3983
.9238
13.485 5.4670
. 15783 -2 5.4353
.6O398 -2 -6.80S3
.10908 -10.371
.24141 -2 12.893
SIGNIF
.OOOO
SIGNIF
.0000
.0000
.0000
.0000
.0000
FORWARD SELECTION
» VAR
1 2.
2 83.
3 81 .
2 2.
3 17.
4 20.
VARIABLE PARTIAL
TEMP IN -.63087
FS'ETW IN .60202
FS-TEMP IN -.39400
TEMP OUT -.05331
ETW IN -.34005
TEMP. 2 IN .45367
SIGNIF
.0000
.0000
.OOOO
.4977
. 0002
.0000
-------
SELECT OPTIONS=STEPWISE,BACKWARD VAR=18;2,20,7,12,17,71,72,74,81 82
MAXIM=100 LEVELS'.05,.10>
Selection of Regression of the Amount (DIFF) by Which Bag 1 CO Exceeds
Bag 1 CO at 75'F for 16 Veh.
ANALYSIS AT STEP 0 (Initial Step) FOR 18.DIFF' N= 119 OUT OF 119
SOURCE DF SUM OF SQRS MEAN SQUARE F-STAT SIGNIF
REGRESSION
ERROR
TOTAL
MULTIPLE R= .87525
VARIABLE
CONSTANT
2. TEMP
20. TEMP. 2
7.PLS
12. FS
17.ETW
71 .PLS-TEMP
72.PLS-TEMP.2
74.FS-PLS
81 .FS*TEMP
82.FS-TEMP .2
ANALYSIS AT STEP 5
SOURCE
REGRESSION
ERROR
TOTAL
MULTIPLE R= .87424
VARIABLE
CONSTANT
2. TEMP
12. FS
17 .ETW
81 .FS-TEMP
82. FS-TEMP. 2
REMAINING
20. TEMP. 2
7.PLS
71 .PLS'TEMP
72.PLS*TEMP.2
74.FS-PLS
10
108
1 18
R-SQR=
PARTIAL
.08594
- .04937
. 00000
. 00000
.27953
- .00545
01299
- . 00000
-.32584
. 18239
(Final
OF
5
1 13
118
R-SOR=
PARTIAL
.34915
.71323
.27547
-.64637
.43369
PARTIAL
-.07237
- .03987
- . 03O09
- .03470
-.-03987
. 1 1430 *6
34905 .
. 14921 +6
.76606 SE=
COEFFICIENT
-152.92
1 .7021
-. 34796 | -2
39.773!
154.42
.10169 - 1
-.62300 -1
. 12302 -2
' -44 .047
-2.2819
. 10687 -1
Step) FOR 18.
SUM OF SORS
. 1 1 4O4 -1-6
35169.
. 14921 -t-6
.76430 SE=
COEFFICIENT
-97.936
.85508
102.01
. 10O15 -1
-1 .8464
.66984 -2
SIGNIF
. 4442
.5736
.7506
. 7 1 4O
.6736
1 1430.
323. 19
17.978
STD ERROR
.59151 +8
1 .3988
. 16507 -1
.59151 +8
.59151 +8
.3361O -2
1 .0998
.91 10O -2
.59151 +8
.63492
.55434 -2
DIFF N= 119
MEAN SQUARE
22808 .
31 1 .23
17 .642
STD ERROR
17 .878
.21589
9.4305
.32876 -2
. 205O5
. 13092 -2
35.366
T-STAT
-.25853 -5
.89639
-.51369
. 57239 -5
.26107 -5
3.0256
-.56645 -1
. 13504
- . 744.66 -6
-3.5939
1 .9278
OUT OF 119
F-STAT
73.283
T-STAT
-5.4781
3 . 9608
10.817
3.0461
-9.0050
5. 1 164
. . 0000
SIGNIF
1 . 0000
.3720
.6085
'• . OOOO
• . oooo
.0031
.9549
.3928
1 . OOOO
.0005
.0565
SIGNIF
.0000
SIGNIF
.OOOO
.OOO1
.0000
.0029
.OOOO
.OOOO
REGRESSION OF T8.0IFF USING BACKWARD SELECTION
STEP R-SQR STD ERROR = VAR VARIABLE
PARTIAL SIGNIF
0
1
2
3
4
5
. 766O6
. 76606
.756O6
.76578
.76553
. 76430
17 .978
17.395
17.314
17 . 744
17.674
17 .542
10
9
3
7
S
5
IN
7.PLS OUT
71.PLS-TEMP OUT
72.PLS-TEMP.2 OUT
74.FS-PLS OUT
20.TEMP.2 OUT
.00000
-.00545
.03467
-.03229
-.07237
1.OOOO
.9547
.7167
.7342
. 4442
-------
5-5
SELECT OPTIONS=STEPWISE,FORWARD VAR=18;2,20,7,12,17,71-74,81-83,91,92,93
MAXIM=100 LEVELS*.05,.10>
Selection of Regression of the Amount (DIFF) by Which Bag 1 CO Exceeds
Bag 1 CO at 75'F for 16 Veh.
ANALYSIS AT STEP 1 (Initial Step) FOR 18.DIFF N= 119 OUT OF 119
SOURCE OF SUM OF SORS MEAN SQUARE F-STAT SIGNIF
REGRESSION
ERROR
TOTAL
1 17
1 18
72307.
76901.
.14921
72307.
657.27
110.01
MULTIPLE R= .69514 R-SOR= .48461 SE = 25.S37
VARIABLE PARTIAL COEFFICIENT STD ERROR T-STAT
CONSTANT
2. TEMP
REMAINING
20. TEMP. 2
7.PLS
12. FS
17.ETW
71 .PLS-TEMP
72.PLS-TEMP.2
73.PLS-ETW
74.FS*PLS
81 . FS-TEMP
82. FS* TEMP. 2
83.FS*ETW
91 .FS*PLS«TEMP
92.FS-PLS-TEMP.
93.FS«°LS*ETW
-.59614
PARTIAL
.27473
-. 10272
.37123
.23127
-.OO670
. 13664
.01750
.33336
. 143O6
. 14778
.41252
. 15279
2 .18404
.38121
73. 145
-.39801
SIGNIF
.0026
.2683
.0000
.0117
.9426
.1401
.3508
.0002
. 1222
. 1 103
.0000
.0986
.0460
.OOOO
5.9946
.85618 -1
12.202
-10.489
ANALYSIS AT STEP 6 (Final Step) FOR 18.DIFF N« 119 OUT OF 119
SOURCE OF SUM OF SORS MEAN SQUARE F-STAT
87.983
REGRESSION
ERROR
TOTAL
4
1 14
1 18
.11270 +6
365O7.
. 14921 +6
28175.
320.23
MULTIPLE R= .86910 R-SOR" .75533 SE- 17.895
VARIABLE PARTIAL COEFFICIENT STD ERROR T-STAT
PARTIAL
SIGNIF
REMAINING
2.TEMP
7 . PLS
12.FS
71.PLS'TEMP
72.r>LS*TEMP . 2
73.PLS-ETW
74.FS-PLS
82.FS-TEMP.2
91'.FS"PLS"TEMP
92 FS*PLS*TEMP.2
93.FS»PLS*£TW
REGRESSION OF 18.DIFF USING FORWARD SELECTION
STEP R-SOR STD ERROR » VAR VARIABLE
. 1 1532
.00387
.01878
.03368
.01504
.OO387
.00519
.17170
.00480
.08610
.00387
. 2 177
.9672
.8421
.7208
.8732
.9672
.9561
.0665
.9594
.3602
.9672
1
2
3
4
5
6
.48461
.57231
.62143
.31485
.€8463
.75533
25.637
23.455
22. 163
22.258
20.228
17.895
.0000
SIGNIF
.0000
.0000
SIGNIF
.0000
SIGNIF
CONSTANT
20. TEMP. 2
17.ETW
81 .FS-TEMP
83.FS-ETW
.47347
-.60221
-.71626
.75492
93.998
.87349 -2
-.46906 -1
-1 . 1526
. 28610 -1
13.003
. 15219 -2
.58239 -2
. 10518
.23278 -2
7.2292
5.7394
-8.0540
-10.959
12.290
.0000
.0000
.0000
.0000
. 0000
PARTIAL SIGNIF
1
2
3
2
3
4
2. TEMP
33.FS-ETW
81 .FS'TEMP
2. TEMP
1 7 . ETW
20. TEMP .2
IN
IN
IN
OUT
IN
IN
- .59614
.41252
-.33888
- . 13071
-.42566
.47347
.0000
.OOOO
.0002
.1601
.0000
.0000
-------
5-6
SELECT OPTIONS = STEPWISE, BACKWARD VAR= 19 ; 2 ,"20 , 7 , 12, 17,71 ,72,74,81 .82
MAXIM=100 LEVELS5-OS,.10>
Selection of Regression of Quotient (RATIO) of Bag 1 CO Divided by
Bag 1 CO at 75'F for 16 Vehicles
ANALYSIS AT STEP 0 (Initial Step) FOR 19.RATIO N= 119 OUT OF 119
SOURCE DF SUM OF SORS MEAN SQUARE r-STAT SIGNIF
REGRESSION
ERROR
TOTAL
MULTIPLE R = .681 12
VARIABLE
CONSTANT
2 .TEMP
20. TEMP. 2
7.PLS
12. FS
17 . ETW
71 . PLS'TEMP
72.PLS*TEMP . 2
74.FS*PLS
81 . FS'TEMP
82.FS«TEMP.2
ANALYSIS AT STEP 7
SOURCE
REGRESSION
ERROR
TOTAL
MULTIPLE R= .57206
VARIABLE
CONSTANT
12. FS
81. FS'TEMP
82.FS-TEMP.2
10
108
1 18
R-SOR=
PARTIAL
- . OOO29
- .00122
. 00000
. 00000
. 1 135O
.01778
- .00563
- . 00000
-. 14079
.08368
(Final
OF
3
115
1 18
H-SOR"
PARTIAL
.56296
-.43346
.29451
891 .60
103O.3
1921 .9
.46392 SE=
COEFFICIENT
-2.8227
-.96814 -3
-.35951 -4
.45237 -1
9.S308
.68549 -3
. 3491 1 - 1
-.91602 -4
-2.3586
- . 16120
.83111 -3
Step) FOR 19.
SUM OF SORS
868.04
1053.8
1921 .9
.45166 SE=
COEFFICIENT
.62845
6.5177
-. 13936
.74O08 -3
89. 160
9.5395
3.0886
STD ERROR
. 10162 +8
.32623
.28360 -2
. 10162 + 8
. 10162 +8
.57742 -3
. 18896
. 15651 -2
. 10162 +8
. 10908
.95238 -3
RATIO N» 1
MEAN SQUARE
289.35
9. 1638
3.0272
STD ERROR
.97537
.89227
.27017 -1
.22393 -3
9 . 3464
T-STAT
-.27776
-.29677
-. 12677
. 44514
.94769
1 . 1872
. 18476
-.58527
- . 23209
-1 .4778
.87267
19 OUT OF
F-STAT
31 .575
T-STAT
.64432
7 . 3046
-5. 1581
3 . 3049
.0000
SIGNIF
-6 1 . OOOC
-2 .9976
-1 .9899
-a 1.0000
-6 1 .0000
.2378
.8538
-1 .9534
-6 1.0000
. 1424
.3848
119
SIGNIF
.OOOO
SIGNIF
.5207
.OOOO
.OOOO
.0013
REMAINING
PARTIAL
SIGNIF
2. TEMP .07341 .4335
20. TEMP. 2 .07109 .4483
7.PLS -.O2253 .3103
17 . ETW . 10982 .2406
71.PLS-TEMP .03515 .7080
72.PLS-TEMP.2 . O488 1 .6028
74.FS*PLS -.02253 .8103
REGRESSION OF 19. RATIO USING BACKWARD
STEP R-SQR STD ERROR. P VAR
0
1
2
3
4
5
6
7
.46392
.46392
. 46392
.46390
.46382
.45879
.45828
.45166
3
3
3
3
3
3
3
3
.0886
.0744
0604
.0467
.0332
.0339
.0220
.0272
10
9
8
7
6
5
4
3
7
2.
20
72 .
71
74.
17
SELECTION
VARIABLE
.PLS
.TEMP
.TEMP. 2
.PLS-TEMP.2
. PLS'TEMP
.FS-PLS
. ETW
IN
OUT
OUT
OUT
OUT
OUT
OUT
OUT
PARTIAL
. 00000
- . OO029
-.OO659
- .01 189
.09647
-.03075
. 10982
SIGNIF
1 . OOOO
.9976
.9450
.9005
.3072
. 7443
.2406
-------
5-7
SELECT OPTIONS=STEPWISE,FORWARD VAR=19;2,20.7,12.17,71-74,81-83,91,92,93
MAXIM*100 LEVELS=.OS,.10>
Selection of Regression of Quotient (RATIO) of Bag 1 CO Divided by
Bag 1 CO at 75'F for 16 Vehicles
ANALYSIS AT STEP 1 (Initial Step) FOR 19.RATIO N= 119 OUT OF 119
SOURCE DF SUM OF SQRS MEAN SQUARE F-STAT SIGNIF
REGRESSION
ERROR
TOTAL
MULTIPLE R» .57659
VARIABLE
CONSTANT
2. TEMP
REMAINING
20. TEMP. 2
7.PLS
12. FS
17. ETW
71 . PLS'TEMP
72.PLS»TEMP.2
73.PLS--ETW
74. FS-PLS
81 . FS-TEMP
82. FS-TEMP . 2
83.FS«ETW
91 . FS"PLS"TEMP
92.FS'PLS"TEMP.2
93.FS"PLS'ETW
ANALYSIS AT STEP 4
SOURCE
REGRESSION
ERROR
TOTAL
MULTIPLE R= .65672
VARIABLE
CONSTANT
2. TEMP
20. TEMP. 2
82.FS-TEMP.2
33.FS-ETW
REMAINING
7.PLS
12. FS
17 . ETW
71 . PLS'TEMP
72. PLS-TEMP . 2
73.PLS*ETW
74. FS-PLS
81 . FS'TEMP
91 .FS-PLS*TEMP
92. FS«PLS*TEMP .2
93 . FS*PLS*ETW
1 638.94 638.94 58.270
117 1282.9 10.965
118 1921 .9
R-SOR= .33246 SE= 3.3114
PARTIAL COEFFICIENT STD ERROR T-STAT
7.9725 .77427 10.297
-.57659 -.84416 -1 . 1 1O59 -1 -7.6335
PARTIAL SIGNIF
.26269 .0041
-.O7387 .4266
.19936 .03O4
. 1 1698 .2071
-.00879 .9247
. 1 1775 . 2041
-.0105O .9102
. 16701 .0707
.08684 .3498
.13610 .1417
.21702 .0183
.08943 .3355
. 16164 .OSO3
. 19046 .0388
(Final Step) FOR 19. RATIO N= 119 OUT OF 119
DF SUM OF SQRS MEAN SQUARE F-STAT
4 828.87 2O7.22 21.613
114 1093.0 9.5877
118 1921.9
R-SOR* .43128 SE* 3.0964
PARTIAL COEFFICIENT STD ERROR T-STAT
7.5243 .1.6640 4.5217
-.4134O -.22931 .47304 -1 -4.8475
.32536 .18132 -2 .49354 -3 3.3738
-.21074 -.34248 -3 .14879 -3 -2.3018
.29133 .68738 -3 .21140 -3 3.2515
PARTIAL SIGNIF
-.06731 .4748
. 17053 .0684
- . 17274 .0649
-.02812 .7654
-.01590 .8661
- . 1 1539 .2194
.05007 .5951
.06203 .5101
-.00961 .9188
-.01590 .8661
-.06731 .4748
.0000
SIGNIF
.OOOO
.OOOO
SIGNIF
.0000
SIGNIF
.0000
.OOOO
.0004
.0232
.0015
REGRESSION OF 19. RATIO USING FORWARD SELECTION
STEP R-SOR STD
1 .33246 3.
2 .37852 3.
3 .40485 3.
4 .43128 3.
ERROR if VAR VARIABLE PARTIAL
3114 1 2. TEMP IN -.57659
2088 2 20. TEMP. 2 IN .26269
1537 3 83.FS-ETW IN .20583
0964 4 82. FS-TEMP .2 IN -.21074
SIGNIF
.OOOO
.0041
.0260
.0232
------- |